Samsung HD204UI/Z4 2tb drive failure?

[jebusau]

i've had an almost identical experience with 5 of these drives i purchased late Oct and RMA'ed them soon after

nothing helpful from the retailer and even less from Samsung.

i finally posted on overclockers AUS (cant post URLs yet) asking for someone to run the test to see if they get the same and was told there is a new version of the tool recently released

i bought two more drives and tested them using the new version 3.01v - no error!
i'm now running the older 3.01p and 3.00g versions to see if the error re-appears

[Shaolin_K]

Thanks for your experiences Jebusau

Two things I need to clarify from the outset:

does your drives:

1. pass the new ESTool version after running the FULL 6+hrs diagnostics (I ask as many will pass the short test - like mine, and as I've mentioned previously, will pass the short test on a loop)

2. exhibit write errors, with the count rate increasing as more and more data is written to the drive (like mine did)?

I would have happily kept the drive if it did not develop WRITE errors (being my main concern; other errors include RAW read rate and G-sense errors).

As I've stated, I was very hesitant in RMA'ing the drive in the first instance, running tests to satisfy myself to avoid the no fault found fee that Scan imposes (not a decision I took lightly).

I've requested a retest last Monday 22nd - a week on and I still haven't got an answer yet. So basically, having took delivery on Oct 22nd - 5 weeks later and still no drive to work with.

[jebusau]

Thanks for your experiences Jebusau

Two things I need to clarify from the outset:

does your drives:

1. pass the new ESTool version after running the FULL 6+hrs diagnostics (I ask as many will pass the short test - like mine, and as I've mentioned previously, will pass the short test on a loop)

2. exhibit write errors, with the count rate increasing as more and more data is written to the drive (like mine did)?

I would have happily kept the drive if it did not develop WRITE errors (being my main concern; other errors include RAW read rate and G-sense errors).

As I've stated, I was very hesitant in RMA'ing the drive in the first instance, running tests to satisfy myself to avoid the no fault found fee that Scan imposes (not a decision I took lightly).

I've requested a retest last Monday 22nd - a week on and I still haven't got an answer yet. So basically, having took delivery on Oct 22nd - 5 weeks later and still no drive to work with.

G'day mate
Yes i've only ever run the full surface scan which takes 6+ hours
and in fact even when you run the short scan, with enough repetitions, the error still comes up
on my initial 5 drives, i ran the full surface scan on each at least twice across three pc's, using v3.01p - it took some days to finish all that up, and the error always came up

Unfortunately i did not start using them or test them further for write errors, i RMA'ed them directly to Samsung Australia since late Oct and have been trying to talk to them about the possibility of an issue with the ES-Tool since then, with no luck

Someone on OCAU told me about the new version of the tool, 3.01V, so i bought two new drives and ran the full surface on them using this version - no errors!

since then i've also ran 3.01P and 3.00G on each of these two new drives and the error follows 3.01P
i'll be monitoring for write errors as you suggest once i start using these two new drives and can update then.

[Shaolin_K]

I don't wish to really make this public, but am now getting really annoyed with the situation right now!

Still no resolution to the matter.

22 Oct 10 - Accepted delivery of the drive and had it for the first week & a half to try convince myself not to go through with RMA process
29 Oct 10 - Requested RMA and collection
04 Nov 10 - Failed collection by DPD
08 Nov 10 - Second failed collection by DPD
12 Nov 10 - Finally collected after wasting time pursuing
22 Nov 10 - Requested a re-test, outlining write errors and how to test
07 Dec 10 - Rang Scan and was advised it was thought the re-test was done on 23 Nov 10 and will follow up
10 Dec 10 - Still no word

As you can imagine my frustration, this is quite ridiculous!

I don't think it's unreasonable at this stage for me to request either a new drive / full refund to close the matter, unless of course Scan could materially substantiate their claims the drive is not faulty with the re-test (demonstrating write error count rate not increasing with further data written to the drive).

I never thought I would end up in this situation

[sammorris]

Just do what I do, ring them repeatedly to make sure they're still doing stuff. You should keep the ethic that something must be happening related to your query every day, and if that isn't immediately obviously the case, ring up and find out what's going on
This approach has helped speed up my RMAs with companies significantly.

[Shaolin_K]

Apparently no fault found on the retest.

Subject to me accepting that there are no faults (per screenshots by Scan this evening when I get home), I guess I'll have to accept the charge.

However, I understand the charge to be £15 + vat, which includes the testing & re-delivery fee.

Scan's trying to charge me £35 + VAT

£10 for collection
£15 for testing
£10 for re-delivery

Apparently the re-test fee of £5 + vat has bee waived.

Is this right?

[jim]

This is why, ridiculously, unless I'm absolutely certain that something will be considered faulty by Scan, I now consider writing it off as a viable alternative.

[codez4me]

Apparently no fault found on the retest.

Subject to me accepting that there are no faults (per screenshots by Scan this evening when I get home), I guess I'll have to accept the charge.

However, I understand the charge to be £15 + vat, which includes the testing & re-delivery fee.

Scan's trying to charge me £35 + VAT

£10 for collection
£15 for testing
£10 for re-delivery

Apparently the re-test fee of £5 + vat has bee waived.

Is this right?

There is a firmware issue with these drives. Samsung have just issued a firmware patch. Can anyone else apply this please and confirm back here if it fixes their problems? Thanks.

I can't post the URL as I haven't made 5 posts, but Google this to find the FAQ and the patch :-

"Patch tool for Identify fail during NCQ write command.(Model: F4EG)"

[rabbid]

This link?
http://www.samsung.com/global/busine...bbs_msg_id=386
File: http://www.samsung.com/global/system...621878F4EG.zip

"If identify commmand is issued from host during NCQ write command in the condition of PC ,
write condition is unstable.
So It can make the loss of written data.
(Model : F4EG HD204UI, HD204UI/Z4, HD204UI/UZ4, HD155UI, HD155UI/Z4, HD155UI/UZ4)
To improve this problem, Please use the attached DOS model patch tool.
1. Please unzip attached file to Bootable media (USB,CD, Floppy...etc).
2. Please connect Target HDD to Primary / Mast.
3. Booting the system to DOS mode by bootable media.(which with this F/W flash program).
4. Please run " F4EG ".
5. When FW Flash successfully message display , please Power Off system, to make F/W flash complete.

Note: Please do not warm boot(Ctrl+Alt+Delet , or push Reset button..etc) your PC during the
F/W patch. Warm boot can make F/W Patch fail and make unexpected problem on HDD."

[JKay6969]

Sorry, in my haste to pass this info on I didn't realise that it had been discovered already, oh well, such is life :-) lol

I have the same problems as you guys/Gals, I have lost about 4TB of Multimedia due to 2x 2TB Samsung HD204UI\Z4's losing partition info, I was lucky, I own 4 of them, It could have been much worse.

I have been scouring the net to find possible solutions and am currently scanning the first drive with ES Tool, it's about 25% through the full surface scan.

I have just lost a RAID 5 ARRAY made up of 4x 2TB Samsung HD204UI\Z4's and 2x 2TB Samsung HD204UI's. The 4x HD204UI\Z4's died but the 2x HD204UI's stayed, completely destroyed the RAID 5 ARRAY as only 1 drive can die to save the data but there was only about 100GB of Multimedia on the ARRAY and that was COPIED over, no loss of data, thankfully.

I decided to make the ARRAY to protect my data from the HD204UI\Z4's affinity to crashing only to have all 4 crash at once, hmm, tell me that I should trust these drives at all. The strange thing is that the non-Z4 drives didn't crash!?! This tells me there is an issue with the Z4 model that causes higher instability.

Anyways, in my searching I happened across a couple of Samsung web pages with firmware updates for the 204UI+UI\Z4 drives, links below.

A37.

If identify commmand is issued from host during NCQ write command in the condition of PC ,
write condition is unstable.
So It can make the loss of written data.
(Model : F4EG HD204UI, HD204UI/Z4, HD204UI/UZ4, HD155UI, HD155UI/Z4, HD155UI/UZ4)

To improve this problem, Please use the attached DOS model patch tool.

1. Please unzip attached file to Bootable media (USB,CD, Floppy...etc).
2. Please connect Target HDD to Primary / Mast.
3. Booting the system to DOS mode by bootable media.(which with this F/W flash program).
4. Please run " F4EG ".
5. When FW Flash successfully message display , please Power Off system, to make F/W flash complete.

Note: Please do not warm boot(Ctrl+Alt+Delet , or push Reset button..etc) your PC during the
F/W patch. Warm boot can make F/W Patch fail and make unexpected problem on HDD.

/global/business/hdd/faqView.do?b2b_bbs_msg_id=386&page=1&faqType=QI&faqKeyword[/url]

A38.

If identify commmand is issued from host during NCQ write command in the Baremarket for PC
and not external cse, write condition is unstable.
So It can make the loss of written data.
(Model : F4EG HD204UI/JP1, HD204UI/JP2)

To improve this problem for /JP1, Please use the attached DOS model patch tool.
*Notice: Please back-up your data in HD204UI before using the tool .

1. Please unzip attached file to Bootable media (USB,CD, Floppy...etc).
2. Please connect Target HDD to Primary / Mast.
3. Booting the system to DOS mode by bootable media.(which with this F/W flash program).
4. Please run " HD204UI_JP ".
5. When FW Flash successfully message display , please Power Off system, to make F/W flash complete.

Note: Please do not warm boot(Ctrl+Alt+Delet , or push Reset button..etc) your PC during the
F/W patch. Warm boot can make F/W Patch fail and make unexpected problem on HDD.

/global/business/hdd/faqView.do?b2b_bbs_msg_id=387&page=1&faqType=QI&faqKeyword[/url]

Both pages have links at the bottom to download a DOS based Firmware update tool that will fix the problems. Each page has a different tool.

It seems I cannot post the links because I'm new here, I took out www and Samsung and com from the links.
I have not had the chance to try the tools yet as I've not completed my full surface scan, but I intend to check them out when it completes. The broken english descriptions seem to describe the problems we're facing so I'm more than a little bit excited about this find.

To the Scan Tech support people, would updating the firmware invalidate the warranty on these drives? I don't want to risk not being able to return them as they are useless in their current state.

Anyone, please let me know how you get on with these new updates.

Regards

JKay6969

[JKay6969]

Apparently no fault found on the retest.

Subject to me accepting that there are no faults (per screenshots by Scan this evening when I get home), I guess I'll have to accept the charge.

However, I understand the charge to be £15 + vat, which includes the testing & re-delivery fee.

Scan's trying to charge me £35 + VAT

£10 for collection
£15 for testing
£10 for re-delivery

Apparently the re-test fee of £5 + vat has bee waived.

Is this right?

This is crazy!!! I know that there is something wrong with my drives, I have NEVER had a drive just lose it's partition info like my TWO HD204UI\Z4's have, The performance from the drives when they are functioning is absolutely terrible, unusable! then to have all FOUR HD204UI\Z4's crash in my RAID ARRAY at the same time!!!

It sounds to me like you have similar data loss issues, how can SCAN possibly think these drives are OK?!?!?

I would like for them to put some important data on these drives and use them for a week, then tell us there's nothing wrong with them after they lose all their data!!!!

I'm gobsmacked that this is your result mate, I'm quite worried now as I've just started the RMA process and don't want to spend another £35+VAT to be told these faulty drives are working OK?!?!

Crazy

[Shaolin_K]

Right, here's an update

First off, thanks to Scan for sorting out my RMA, albeit with some stress & frustration over the past 6 weeks!

To my pleasant surprise, a replacement drive was waiting for me when I got home on Friday evening after work - I think it was picked & despatched on Dec 15th. I had no SMS from DPD re. delivery, no email confirmations from Scan or any updates on the RMA status page on Scan's website (checked Dec 14th to 17th).

After a quick test using Samsung's ESTool (3.01v), followed by the full surface scan, I'm glad to report that the drive did not report the AJ41 error code which I encountered with the RMA'd drive. However, I must make clear that I did not RMA the drive solely on this error, as it did occur to me that v3.01p (which was used then) could have been incompatible.

My main gripe was with the increasing write errors which developed as more data was written to the drive. It has now been identified as an error with the drive's firmware, hence the release of the patch. Had I 'd known about the issues affecting this drive, I wouldn't have purchased it in the first instance, and certainly would not have RMA'd it (requested Oct 29th) knowing there may be a fix on the horizon (which I believe was released around Dec 10th?).

Regarding my replacement drive, I've formatted it, copied data over, re-formatted and copied more data to it (in total around ~100gb) and I'm glad to report ZERO write errors - now I'm more confident with using the drive for its intended purpose. As mentioned previously, none of my other drives exhibited write errors - all report 0 under SMART.

The only aspect I'm not happy with with the replacement is the Crystal Mark benchmark. After formatting (4k alignment) under Win 7, the RMA'd drive showed results similar to others around the net, whereas my current one performs a little lower than expected (eg. sequential read ~140mb vs ~125mb and write ~138mb vs 120mb). This is on a fresh quick format. What's interesting is the results are a little higher when there's around 60gb of data on the drive (it went down on the other drive in this scenario).

Also, the drive is not recognised as a Samsung Advanced Format drive using their alignment tool?! Anyone else with this problem?

With Crystal Mark results being lower, NIL write errors after testing, the replacement drive being what appears to be from the same batch as that RMA'd (drive manufactured Sept 2010) and same firmware number, I can only conclude Scan may have applied the patch before sending me the replacement? [SCAN - can you confirm?]

In any case, I think I'll leave this where it is - I'll live with the slightly lower performance.

If I'd puchased this drive now (approx 2 months down the line), I'd have saved around £5 - £7, been frustration-free and (may) have known about the write issues.


edit - I only had the drives under IDE mode, but I've been reading about the identify device bug and understand it only arises in certain setups such as RAID / NAS, with silent data loss?

[Shaolin_K]

Has anyone applied the patch yet - if so, are your benchmarks the same?

[codez4me]

Has anyone applied the patch yet - if so, are your benchmarks the same?

+1 to the above. Have just purchased an external drive which uses 2 of these drives so wondering if flashing them fixes the write issues. Avoiding using it at the moment!

[dangel]

I might be adding my tuppence here - my media centre died overnight and the kingston SSD appears to be OK (mounted in a usb caddy and backed up with acronis over breakfast) so that leaves the sammy. The system is saying boot device missing and i found it locked up during boot (win7 boot animation playing) this morning.

I've ordered a new drive (I need a second anyway) but i'll struggle to have time to do any real testing tonight - however i did get a warning from windows a few days ago that hard drive failure was imminent (first time i've seen that) I'm dreading the RMA as my experience (scan) hasn't been too clever in the past. Hope springs eternal though and i'll keep you up to date on my experiences on both the hardware and Scan RMA front.

Still it's odd, since the boot drive is the SSD not the HD and the 2TB drive went in many months after installation as a replacement for a 1TB. Unless I forgot about having to reassign the boot manager... memory doesn't serve me well but i'll soon sort it.

FYI a few days BEFORE the warning in windows i'd updated the firmware I have both drives running off the intel controller (i think in RAID mode but i'll check),.

The kicker is i've got an USB 3.0 caddy and a USB 3.0 card (on the way) to do my backups.. I think i have a partial backup of my data on a USB drive from dec (took 17 hours) but i'll have lost anything since.

[dangel]

Update: put the drive in another machine late last night - no partitions at all!

Drive was made 08.2010 Arse.