arga

Samsung 1TB HD103UJ: HDDScan problem (476 green blocks)

Recommended Posts

Hdparm has been around for i think 15-20 years and it is open source... and it gets installed in all linux versions by default

and does it work with modern hard drives??

Share this post


Link to post
Share on other sites

@6_6_6: Yes that's that function. The article is from 2006 and I think its new marketing name is now "Virtual DualBIOS" or "Virtual QuadBIOS".

HDparm is developed actively and works with modern harddrives. At least the linux version.

The MBR can be restored with booting the Windows install CD and typing fixmbr on the recovery console.

The problem with that Gigabyte feature is that it destroys data at the end of the drive without asking. Imagine a GUID-partitioned drive for example...

Share this post


Link to post
Share on other sites
@6_6_6: Yes that's that function. The article is from 2006 and I think its new marketing name is now "Virtual DualBIOS" or "Virtual QuadBIOS".

weaker, QuadBIOS and Xpress Recovery are not the same. Xpress Recovery = Ghost.

In order to be able to boot into HPA/DCO data, i think Gigabyte is setting the HPA as the entire drive, but failing to restore it when done. From what i read, that is how other software utilize HPA. This is more of an Xpress Recovery thing than a BIOS backup since BIOS backup does not need to be boot into. It is just a file that needs to be read.

arqa, did you run Xpress Recovery before on this drive or on this system? It needs to be explicitly installed.

Share this post


Link to post
Share on other sites

@both:

I think:

Xpress BIOS Rescue=Vitual BIOS

Xpress Recovery=Ghost

No, I didn't install nor run any of these.

@666:

Is the MBR restored automatically when you Init, partition, and Format the drive with Windows?

Edited by arga

Share this post


Link to post
Share on other sites

Surprise:

I got a bad sector in my good 1TB drive!!

Until now, the drive was empty, and formatted, and scans were good.

But today, I have filled the drive with data, and now a Bad sector appears!!

The scan is good (8 greens) but now there is a bad block.

The bad block appears in both MHDD and HDDScan!! How is this possible??

I did nothing out of the ordinary on this drive. It was in the other one, where I did all

the experiments.

On the other side, I have also filled the bad 1TB drive with data, and the scan has improved a lot. Now only 49 greens in MHDD. And no bads.

Edited by arga

Share this post


Link to post
Share on other sites
@666:

Is the MBR restored automatically when you Init, partition, and Format the drive with Windows?

Yes. Dealing with so many internal ATA issues, i was thinking more of a firmware when i said i had no idea. Yeas, partition / format fixes that. It just contains bootloader information (plus partition information) or if it is not the OS disk, just partition information.

Share this post


Link to post
Share on other sites

Well... Did you pay attention when i said: "arga, it is not about what commands are being issued. You are playing with lots of tools that access drive's internals. A bug in any of them about your controller, your BIOS, your drive, etc... Or a bug in the drive's firmware... Well, if you are fiddling too much, you surely are likely to break it."

You might not have performed anything on this drive, but this drive might have been powered on with the other drive... and some of hte utilities might have accessed it. Besides, you ran HDDScan many times on this drive. Therefore you had tools accessing it.

Bad sector might appear because a drive does not mark a bad sector as BAD when a read is performed on it hoping that some utility would be able to recover data from that bad sector. But it marks the sector in Pending Allocation Sectors in SMART. Now when a write is performed on that sector (format/erase/copy, etc), it automatically remaps it as bad (if it cannot recover it).

No big deal, just full format once more and see if remapped sectors count goes up. If not, just use it. Modern drives come with plenty of bad sectors (p-list) and they have spares to accommodate new bad sectors (g-list). Few bad sectors here there is not a problem. I remember when the drives had sector map stickers on them that we had to enter manually when we low-level formatted them :)

Surprise:

I got a bad sector in my good 1TB drive!!

Until now, the drive was empty, and formatted, and scans were good.

But today, I have filled the drive with data, and now a Bad sector appears!!

The scan is good (8 greens) but now there is a bad block.

The bad block appears in both MHDD and HDDScan!! How is this possible??

I did nothing out of the ordinary on this drive. It was in the other one, where I did all

the experiments.

On the other side, I have also filled the bad 1TB drive with data, and the scan has improved a lot. Now only 49 greens in MHDD. And no bads.

Share this post


Link to post
Share on other sites
Well... Did you pay attention when i said: "arga, it is not about what commands are being issued. You are playing with lots of tools that access drive's internals. A bug in any of them about your controller, your BIOS, your drive, etc... Or a bug in the drive's firmware... Well, if you are fiddling too much, you surely are likely to break it."

You might not have performed anything on this drive, but this drive might have been powered on with the other drive... and some of hte utilities might have accessed it. Besides, you ran HDDScan many times on this drive. Therefore you had tools accessing it.

Bad sector might appear because a drive does not mark a bad sector as BAD when a read is performed on it hoping that some utility would be able to recover data from that bad sector. But it marks the sector in Pending Allocation Sectors in SMART. Now when a write is performed on that sector (format/erase/copy, etc), it automatically remaps it as bad (if it cannot recover it).

No big deal, just full format once more and see if remapped sectors count goes up. If not, just use it. Modern drives come with plenty of bad sectors (p-list) and they have spares to accommodate new bad sectors (g-list). Few bad sectors here there is not a problem. I remember when the drives had sector map stickers on them that we had to enter manually when we low-level formatted them :)

Well, I don't know, I did read the SMART report several times when I ran HDDScan, and there was no pending allocation. Everything was ok. I saved these reports, and I have them with me.

The bad sector appeared suddenly after I filled the drive with data, and I ran MHDD. This was the first time I wrote data on this drive. Until then, it was only formatted

Yes, I powered this drive at the same time with the other one. In fact, after I filled the other one with data, I copied that data to this one.

I have ran Samsung's ESTOOL, and indeed it detects the problem: it reports 6 consecutive bad LBAs. It prompts me to ERASE the drive, and in case the error persists, contact Samsung.

So, isn't a bad block enough to return the drive? This one is less than 2 months old!!

Share this post


Link to post
Share on other sites

Just erase the drive and see what comes up. Watch for more bad sectors, if there is an increase, just return the drive. If no increase, i believe it is okay. Make sure you make md5 sums of your files. So you can just run md5 verify to test data integrity.

I have ran Samsung's ESTOOL, and indeed it detects the problem: it reports 6 consecutive bad LBAs. It prompts me to ERASE the drive, and in case the error persists, contact Samsung.

So, isn't a bad block enough to return the drive? This one is less than 2 months old!!

Share this post


Link to post
Share on other sites

md5? but in theory there is no need for that. The normal filesystem already has built in redundancy to detect errors, no? in fact, HDDScan and MHDD detect Bad Sectors without the need of md5 or any other extra redundancy. Just that they don't tell you what files are affected.

Isn't there any program capable of verifying the files, using this built-in redundancy?

---

Is there any way to disable Gigabyte's Virtual DualBIOS backup? After what has happened, I don't feel safe with it. I would gladly disable it.

But I don't see anything about it on BIOS or in the manual.

-----

re. the bad sectors that I got, I don't know if should erase the disk to force reallocation, because once reallocated, the ESTOOL will pass without error, and thus, the shop wouldn't want to RMA... this is all very annoying grrr.

Share this post


Link to post
Share on other sites

You said there were bad sectors on the old disk. And you moved your data there. Then copied to the other disk that did not have bad sectors. Were the sectors reallocated before or after the move? Which one of your files is corrupt now? And which ones are not? Why to care about HDDScan or MHDD? No bad sectors found but file system corruption. Then what? Isn't it the files that ultimately hold your data and give a meaning to sectors?

I do have an md5 sum of all the files on a drive. And that is archived offline somewhere else. Any time a large copy/move operation is performed on a drive, i verify md5 checksums. Corruption, utilities modifying the drive... does not mean anything to me for as long as my md5 sums check fine. I don't bother that much with drive utilities except occasional health checks. I run them if there is a problem with my checksums. And any time i run a disk utility, whether there is a problem with the drive or not, i verify checksums before taking the drive offline.

Copy a file from a drive to another. Halfway through the copy, power down the system. Reboot. Now you have two identical files. And there is no file system corruption. But one of them is corrupt. Filesystem has no way of verifying file contents. md5 sum does.

No idea about DualBIOS.

Erasing the disk won't force reallocation of bad sectors if "Pending Reallocation" count is empty. Bad sectors are already marked. It will just check if there are more bad sectors. Secure erase will try to overwrite the bad sectors to avoid data extraction from g-list, but i don't think firmware will free them. They will still be marked as bad.

If you dont get more bad sectors, keep using the drive and watch it for a while. I have plenty of drives with bad sectors.

md5? but in theory there is no need for that. The normal filesystem already has built in redundancy to detect errors, no? in fact, HDDScan and MHDD detect Bad Sectors without the need of md5 or any other extra redundancy. Just that they don't tell you what files are affected.

Isn't there any program capable of verifying the files, using this built-in redundancy?

---

Is there any way to disable Gigabyte's Virtual DualBIOS backup? After what has happened, I don't feel safe with it. I would gladly disable it.

But I don't see anything about it on BIOS or in the manual.

-----

re. the bad sectors that I got, I don't know if should erase the disk to force reallocation, because once reallocated, the ESTOOL will pass without error, and thus, the shop wouldn't want to RMA... this is all very annoying grrr.

Share this post


Link to post
Share on other sites

@6_6_6: arga is right, I think you confused Xpress BIOS Rescue and Xpress Recovery (v1 or v2). The term Xpress BIOS Rescue is no longer in use (to my knowledge) and Xpress Recovery never really took off.

At the moment there is no way to disable Dual/Quad Virtual BIOS. Best you can do is to stay with the latest BIOS version (that has got the fix). I agree that it would be very good if it was choosable for the user if he wants it or not.

Share this post


Link to post
Share on other sites

okay, I see that md5 redundancy can solve many other situations when a file is broken.

I was referring to bad sectors. Apparently Windows can detect that a file is broken when the cause is a bad sector. without having any other redundancy.

I am going to explain the precise sequence of what happened to my F1 (1TB) drives.

Drive 1: Purchased in July. Initially good scans.

Drive 2: Purchased in August. Initially horrible scans, lots of greens, but no bad blocks.

Lots of tests but no writting on the disks...

...

Now, first I filled Drive 2 with data. Scan improved. and no bad blocks.

Then I copied Drive 2 contents into Drive 1 (just Copy in Windows). And then, I run MHDD, and BadBlock appears in Drive 1.

I have just checked SMART in this drive 1: it says: "6 pending sectors". No less than 6 sectors!

They are still not reallocated. Just pending for reallocation.

If I erase them with ESTOOL, I think it will map them out, and the drive will be "ok" again.

You said there were bad sectors on the old disk. And you moved your data there. Then copied to the other disk that did not have bad sectors. Were the sectors reallocated before or after the move? Which one of your files is corrupt now? And which ones are not? Why to care about HDDScan or MHDD? No bad sectors found but file system corruption. Then what? Isn't it the files that ultimately hold your data and give a meaning to sectors?

I do have an md5 sum of all the files on a drive. And that is archived offline somewhere else. Any time a large copy/move operation is performed on a drive, i verify md5 checksums. Corruption, utilities modifying the drive... does not mean anything to me for as long as my md5 sums check fine. I don't bother that much with drive utilities except occasional health checks. I run them if there is a problem with my checksums. And any time i run a disk utility, whether there is a problem with the drive or not, i verify checksums before taking the drive offline.

Copy a file from a drive to another. Halfway through the copy, power down the system. Reboot. Now you have two identical files. And there is no file system corruption. But one of them is corrupt. Filesystem has no way of verifying file contents. md5 sum does.

No idea about DualBIOS.

Erasing the disk won't force reallocation of bad sectors if "Pending Reallocation" count is empty. Bad sectors are already marked. It will just check if there are more bad sectors. Secure erase will try to overwrite the bad sectors to avoid data extraction from g-list, but i don't think firmware will free them. They will still be marked as bad.

If you dont get more bad sectors, keep using the drive and watch it for a while. I have plenty of drives with bad sectors.

Share this post


Link to post
Share on other sites

Hey, I know it sounds ridiculous, but could it be that my bad sectors happened because my drive is mounted on a plastic adaptor, without electric contact with the chassis?

The bad sectors appeared when I did the massive writing on the disk.

Look at this:

http://users.iafrica.com/c/cq/cquirke/baddata.htm

Pattern: Snakebite

Some flaky hardware issues can cause odd bytes of data to get bent - perhaps one quantum out of a billion or so, which translates to steady bit-rot over months of use. Most of the time this will affect what is written to disk, but once in a while it may affect where a sector is written to disk (and therefore cause more obvious damage). As PCI, processor registers and RAM busses work with at least 32 bits at a time, the corruption will typically take the form of 4 consecutive bad bytes.

I've seen this happen with perfectly healthy hardware for no reason other than that the hard drive shell had no metal to metal grounding contact with the PC case; a problem I was able to reproduce. For this reason, I recommend grounding the chassis of hard drives to case when doing casual data transfers or using removable drive brackets.

I don't understand well what it says, but could be related.

Share this post


Link to post
Share on other sites

I think these 6 Bad sectors are not bad sectors really.

Is there any way to free them up?

I have been googling, and nothing conclusive, but it seems that you can "unmark the bad blocks" and get them back for use, by simply deleting the MBR, or repartitioning the drive. Is that ok?

well, maybe just erasing them, will do the trick, as they are still "pending", and "not yet reallocated"...

Share this post


Link to post
Share on other sites
I think these 6 Bad sectors are not bad sectors really.

Is there any way to free them up?

I have been googling, and nothing conclusive, but it seems that you can "unmark the bad blocks" and get them back for use, by simply deleting the MBR, or repartitioning the drive. Is that ok?

well, maybe just erasing them, will do the trick, as they are still "pending", and "not yet reallocated"...

Okay, I will explain this once more arqa:

They are faulty sectors. Something (OS/controller/software or whatever) tried to READ them... but it could not succeed. The firmware of drive noticed this and marked them as PENDING (BAD). They are bad, but they are marked as PENDING. The reason for this is: The drive's firmware thinks that some other tool/software might be able to extract data from those bad sectors if they are left untouched and for this reason, it is wiser not to remap them at this stage. Remapping them will make those PENDING sectors inaccessible to other tools.

But when something (OS/controller/software/erase/format or whatever) tries to write these sectors, the firmware will recheck again and if a WRITE succeeds, it will be removed from PENDING and marked as good (REALLOCATED count will not change since they are good after all). But if a WRITE fails, the firmware will mark them as BAD for good and remap them. They will be removed from PENDING and moved to REALLOCATED.

And no, you cannot free bad sectors. That is remapped by drive's firmware. What you are refering with umarking bad blocks via partition /mbr is if the drive's all spare sectors have been used and OS starts finding bad sectors and start remapping them since firmware could not do anything about it (reallocation table full). In this case, the drive is already toast. Deleting / partitioning... well, dont bother. Drive is gone.

All my drives (even those that are insulated/wrapped to eliminate noise) are properly grounded. So I woudn't know about that link. But i had drives suspended in rubber bands with no grounding and i had no problems with these. Still, it wouldn't hurt to properly ground the drive.

You must ultimately decide what you want to do with this drive. RMA or use. My advice would be: Format and watch for reallocated, pending, unc. If no changes for a while in the numbers, keep using the drive, the drive is fine.

@weaker:

No, i am not confused. Disk imaging software is not something to confuse with BIOS utilities. And that link i gave saying Gigabyte Xpress Recovery uses HPA is from the author of a low-level hard disk utility. I reckon he would know what he is saying.

Share this post


Link to post
Share on other sites

@6_6_6: What I try to tell you is that the function called Virtual Dual BIOS also uses HPA to store the BIOS image file. I never denied that XPress Recovery uses HPA, I know that it does, but Virtual BIOS does, too.

Share this post


Link to post
Share on other sites

@666:

Ok, thanks for the detailed explanation about bad sectors. I will format this drive, and see if the pending bads have been unmarked. But first I will ground the drive.

@weaker: yes, 666 suggested that my disk's shrinkage to 32MB was caused by Gigabyte Ghosting function, but I agree with you, I think the cause was the Bios backup function.

I think the Gigabyte Ghosting function doesn't reserve any space in disk unless you invoke this function... but I have no confirmation of this, because I haven't tested it.

Share this post


Link to post
Share on other sites

AFAIK the Gigabyte ghosting function uses a partition hidden by HPA but I also don't know exactly how it works. It was advertised as a must-have but wasn't supported very good later on. And I didn't hear about users using it often.

Share this post


Link to post
Share on other sites

sorry for the delay:

I have formatted the drive using Windows. Long format.

But nothing has changed. In SMART, the sectors are still marked as Bad, but only pending,

and in HDDScan they show up when I do a partial scan around the affected area.

Proceeding to full scan.

Share this post


Link to post
Share on other sites

Run Manufacturer's disk utility and do an erase there. Sometimes that is required to remap.

From smartctl FAQs:

"If your disk has bad sectors (for example, as revealed by running self-tests with smartmontools) and the disk is not able to recover the data from those sectors, then the disk will not automatically reallocate those damaged sectors from its set of spare sectors, because forcing the reallocation to take place may entail some loss of data. Because the commands that force such reallocation are Vendor Specific, most manufactuers provide a utility for this purpose. "

sorry for the delay:

I have formatted the drive using Windows. Long format.

But nothing has changed. In SMART, the sectors are still marked as Bad, but only pending,

and in HDDScan they show up when I do a partial scan around the affected area.

Proceeding to full scan.

Edited by 6_6_6

Share this post


Link to post
Share on other sites

okay, understood.

Anyway, before running ESTOOL>erase I will try to erase these sectors with HDDScan (this way they won't be reallocated yet).

Now I am still doing the full scan to see if there are more bads.

Something I don't understand, is why this disk didn't develop Bad Sectors when it was formatted first time (long format, writing to disk), but developed 6 bad sectors when I filled it with data. :blink:

Run Manufacturer's disk utility and do an erase there. Sometimes that is required to remap.

From smartctl FAQs:

"If your disk has bad sectors (for example, as revealed by running self-tests with smartmontools) and the disk is not able to recover the data from those sectors, then the disk will not automatically reallocate those damaged sectors from its set of spare sectors, because forcing the reallocation to take place may entail some loss of data. Because the commands that force such reallocation are Vendor Specific, most manufactuers provide a utility for this purpose. "

Edited by arga

Share this post


Link to post
Share on other sites

Hi:

sorry for the delay.

here is the update:

The Bad Blocks were unmarked after I erased the drive with HDDScan.

Now this drive seems ok.

The other drive, the one with many "greens", remains the same, now about 60 greens.

I will wait a couple of weeks, and scan again, to see if the number of greens rise again.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now