Samsung 2TB green drives defective firmware?

larrymoencurly said:
It leaves the data intact, as did the firmware upgrades for my Hitachi 7K1000.C and Seagate 7200.12.
Click to expand...

I can confirm this, although I'd suggest backing up important stuff just to be sure.

Try using a separate computer to do the updates if you use a RAID controller - or at least connect them to the internal controller. I did 2 disks first and then confirmed that my raidz2 worked before doing the rest of my 10 disks.

I am really sorry for the thousands of people owning this disk and not knowing about this serious fault. What makes me more sad is that finding this and that a fix is available makes the drive even more attractive to me now (I've been scratching my head about this issue for weeks - blaming OS, controllers, PSUs, ram and whatnot)... Usually when 10 disks all fail it's something else that is to blame - but not in this case...
 
Thanks.

My last post wasn't clear. I have a software RAID using mdadm and I'm not using a RAID controller so all drives are connected directly to the motherboard controller. I was just hoping the firmware would update all drives at the same time and I wouldn't have to power the computer on and off lots of times and swap ports.

On my question regarding the firmware file changing come to think of it I downloaded the update last night. I will do a compare with the update that is currently on Samsung's website tonight and see if it has changed / been renamed / it's the same and I'm just imagining things!
 
Flintstone said:
I am really sorry for the thousands of people owning this disk and not knowing about this serious fault. What makes me more sad is that finding this and that a fix is available makes the drive even more attractive to me now
Click to expand...

I agree, this is really bad especially for those who have not heard about this. The big problem is SMART is not the only way to trigger the random chance of data loss.

(I've been scratching my head about this issue for weeks - blaming OS, controllers, PSUs, ram and whatnot)... Usually when 10 disks all fail it's something else that is to blame - but not in this case...
Click to expand...

I saw this first in October when I tested my second F4 (I did not test the first one). I thought it was a bug in badblocks or the linux kernel or some damage to my motherboard. Although I was confused because I have never seen this behavior in the 2 years the machine has been running and none of my other drives were corrupting.
 
drescherjm said:
I thought it was a bug in badblocks or the linux kernel or some damage to my motherboard. Although I was confused because I have never seen this behavior in the 2 years the machine has been running and none of my other drives were corrupting.
Click to expand...

In the last years it's been a radical shift when it comes to storage. 4K disks (with 512k emulation) and SSD disks (with TRIM) require extensive knowledge to perform up to spec. Not to mention WD with its mess of TLER and WDIDLE3 with various support.

Seems like today it's lies, damned lies and harddrive specs. WD still says their green disks are variable speed - but I haven't seen ANYONE confirm this.

Now we have to patch all harddrives for bugs aswell. It's a MESS!
 
Well to answer my own post Samsung have just renamed the file. The one I downloaded yesterday was named 813811HD204UI.EXE and the new one HD204UI.EXE. Both files are identical.
 
I wonder if that is different from the one that was originally released. I mean 181550HD204UI.EXE
 
utar said:
My last post wasn't clear. I have a software RAID using mdadm and I'm not using a RAID controller so all drives are connected directly to the motherboard controller. I was just hoping the firmware would update all drives at the same time and I wouldn't have to power the computer on and off lots of times and swap ports.
!
Click to expand...

Hey, would you mind confirming if the flash utility upgraded all your drives at the same time?
 
drescherjm said:
I wonder if that is different from the one that was originally released. I mean 181550HD204UI.EXE
Click to expand...

utar already mentioned they are identical (I can confirm). Probably just renamed it so that people have to deal with the 8.3 dos file name weirdness.
 
utar said:
I was just hoping the firmware would update all drives at the same time and I wouldn't have to power the computer on and off lots of times and swap ports.
Click to expand...

I didn't try doing more than one drive at a time, but I did 5 drives one-at-a-time via an external eSATA dock and can confirm that you can swap drives without power cycling the computer. I would just swap the drive and then rerun the update utility from the DOS prompt. The utility rescans the SATA ports and reports the serial number of the drive, so I was able to confirm that it matched the drive I just inserted in the dock.

In case it matters, my motherboard uses one of the Intel ICH10 ports for eSATA, and I had AHCI enabled in the BIOS.
 
FuSoYa said:
utar already mentioned they are identical (I can confirm). Probably just renamed it so that people have to deal with the 8.3 dos file name weirdness.
Click to expand...

The reason I questioned that utar mentioned a different filename. 813811HD204UI.EXE versus 181550HD204UI.EXE
 
Meh. FreeDOS wouldn't complete the boot sequence until I plugged in a CD/DVD drive.
 
xnoodle said:
Meh. FreeDOS wouldn't complete the boot sequence until I plugged in a CD/DVD drive.
Click to expand...

Did your freedos have a menu that asked you about this? The one I installed asked me if I wanted cd/dvd support during boot as well as highmem support.
 
Metaluna said:
I didn't try doing more than one drive at a time, but I did 5 drives one-at-a-time via an external eSATA dock and can confirm that you can swap drives without power cycling the computer. I would just swap the drive and then rerun the update utility from the DOS prompt. The utility rescans the SATA ports and reports the serial number of the drive, so I was able to confirm that it matched the drive I just inserted in the dock.

In case it matters, my motherboard uses one of the Intel ICH10 ports for eSATA, and I had AHCI enabled in the BIOS.
Click to expand...

I wish my eSATA dock would work with this update :(
 
john4200 said:
For anyone having trouble creating a bootable FreeDOS USB flash drive, you might try UNetbootin:

http://unetbootin.sourceforge.net/

It can create bootable USB flash drives for a lot of different things, and one of the choices is FreeDOS.

After you create the FreeDOS flash drive, just copy the Samsung exectuable to the root directory of the flash drive. When you boot from FreeDOS, you will need to type C: to get to the Samsung executable that you copied.
Click to expand...

I tried the UNetbootin method as mentioned above and it worked perfect. As John mentioned above, as soon as the machine boots into DOS you will be at the A:> prompt. You need to type C: to get to the C:> prompt and then type the name of the Samsung executable and it will flash your hard drive the with new firmware.
 
I am correct in assuming HD203WI drives are not impacted by this?
 
I applied the F4EG.exe firmware upgrade to the 8 HD204UI drives I am trying to build a RAID 5 array with. I still get the problem described in this thread. It upgraded the firmware in the drives to 1AQ10003 not 1AQ10001. I am using an Adaptec 3085 cobtroller with the drives in an enclosure that allows the use of SATA drives as Serial Attached SCSI drives. The array gets created OK and during the Build and Verify, one or more of the drives drop out. It seems to be random when and which drive(s) drop. I'm beginning to think I am going to have to return the 20 drives I bought, I was planning on building 2 of these Arrays (I have two enclosures and two controllers) so I purchased 20 drives (4 for spares). When I put the original 8 1TB drive back in the enclosure, it all works fine.
 
Last edited:
The firmware update does not change the version number. Were your drives version 10003 before you ran the update?
 
No they were 1AQ10001 before the upgrade. When I applied one of the other upgrades (089312HD240UI_JP) it said that it was ugrading to 1AQ10001. That didn't work either, so I applied the F4EG upgrade which raised the version to 1AQ10003. That didn't fix the problem either. I'm wondering if a different controller would help. I would think the Adaptec 3085 would be capable though.
 
On my last of 3 F4s, I did the firmware update yesterday with the program found at faq 37 then and my firmware revision stayed at 1AQ10001.

Code: 
jmd0 ~ # smartctl --all /dev/sdd
smartctl 5.40 2010-10-16 r3189 [x86_64-pc-linux-gnu] (local build)
Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net

=== START OF INFORMATION SECTION ===
Model Family:     SAMSUNG SpinPoint F4 EG series
Device Model:     SAMSUNG HD204UI
Serial Number:    S2HGJDWZ806049
Firmware Version: 1AQ10001
User Capacity:    2,000,398,934,016 bytes
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   8
ATA Standard is:  ATA-8-ACS revision 6
Local Time is:    Sun Dec 19 12:39:14 2010 EST
 
When I applied this upgrade (FAQ 37) - F4EG.exe the software reported that the version number was now 1AQ10003. I ain't making this s..t up. It's for real. It is really strange that some people are reporting that the version number stays at 1AQ10001. Strange.
 
It may be that they changed the firmware again. There seem to be several versions of this update..
 
Has anybody had any success using these drives in a SAS SATA array with any SAS controller?
 
I'm using the Adaptec 3085 that has 2 external SFF8088 SAS connectors. I would consider switching to another controller if it would solve the problem, but it would have to expose 2 SFF 8088 ports. I'm having a lot of trouble trusting these drives with my data.
 
mevenson said:
It makes me wonder if Samsung even has a clue.
Click to expand...

drescherjm said:
It may be that they changed the firmware again. There seem to be several versions of this update..
Click to expand...

mevenson said:
That defintely sounds more plausible.
Click to expand...

No it doesn't.

If there's been two BIOS updates in under a week for this drive, I'm going with your first instinct.

If they're changing the firmware twice in a week, their customers should not be left with uncertain statements like "it may be" and "seem to be".

I've never had to do a HD firmware update in the decades of experience I've had with building machines. Samsung should know leaving uncertainty surrounding this does their customers no good whatsoever.

Clueless? More like careless. I've bought seven of these drives, and it sounds like it's time to consider if there will be an eighth. Fool me once, shame on you. But fool me twice? F U!

Nothing about something like this should be left uncertain to your customers if you care about them.

If this firmware bug caused a "payment hole" where our payments for their drives sometimes just didn't get deposited into Samsung's pockets, you know they'd have that fixed yesterday with absolute certainty, and would make it crystal clear what was going on, etc etc.

If people who already updated need to update again if there's a newer firmware, Samsung needs to make this absolutely crystal clear. Releasing a new firmware to fix a write hole issue, then quietly releasing an even newer one days later does not give me much confidence that the one I just updated to that's no longer available is a good firmware to be using. What if they updated the firmware and found an "oh f-ing shit!" bug in it, and thanks to embarrassment silently released another fix, but thanks to being embarrassed to admit it, they didn't state to update again, and so people testing it and thinking it was okay did not?


If your car was recalled, then fixed, then a second silent recall was done post-fix because of an issue with the fixed part, wouldn't you want to know about it?

Again: not clueless; careless. Time to return to WD quality, me thinks.
 
It appears the first and last executable are the exact same file.

Code: 
john@jmd0 ~/Downloads/temp $ md5sum -b F4EG.EXE ../181550HD204UI.EXE
ad1930d9fe53a6c8e20c3f57936452d1 *F4EG.EXE
ad1930d9fe53a6c8e20c3f57936452d1 *../181550HD204UI.EXE
john@jmd0 ~/Downloads/temp $ shasum -b F4EG.EXE ../181550HD204UI.EXE
1c9b6ca1e83c916ae27ec9bf326e6d94c11059d4 *F4EG.EXE
1c9b6ca1e83c916ae27ec9bf326e6d94c11059d4 *../181550HD204UI.EXE
 
drescherjm:

I get the same md5sum as you, for one I downloaded on Dec 11, and the one I downloaded today, Dec 19:

$ md5sum -b dec11.EXE dec19.EXE
ad1930d9fe53a6c8e20c3f57936452d1 *dec11.EXE
ad1930d9fe53a6c8e20c3f57936452d1 *dec19.EXE
 
Also, it is hard to tell from just examining the strings, but it does look like the FW version contains a 1AQ something, maybe ending in 00001:

Code: 
$ strings F4EG.EXE | grep -i -C 10 fwver
`mMD
\`FE
x1	K
DHx`p(D
y(D%qMy(Deq
>p	x
p3x;p#F
F.p%x[
:|7T`[
ActiveFW : 
FWVer : 
FLASHDAT
1AQ3c0p_
.163
00001
GEO_00
>DDpI
^P_@`
b)cCd
ebfegfh
t5u@v
 
Here's what I get when I do an MD5 on the files I used:

G:\install\Samsung HDD stuff\089312HD240UI_JP>md5 HD240UI_JP.EXE
EEE0492525D836392551021BB744CDCB HD240UI_JP.EXE

G:\install\Samsung HDD stuff\621878F4EG>md5 F4EG.EXE
AD1930D9FE53A6C8E20C3F57936452D1 F4EG.EXE
 
drescherjm said:
It may be that they changed the firmware again. There seem to be several versions of this update..
Click to expand...

The date on F4EG.exe inside the zip file is Dec 14.


Does anyone have a copy of this file from a while back (or have any idea when it got posted)?
 
mevenson said:
Here's what I get when I do an MD5 on the files I used:

G:\install\Samsung HDD stuff\089312HD240UI_JP>md5 HD240UI_JP.EXE
EEE0492525D836392551021BB744CDCB HD240UI_JP.EXE

G:\install\Samsung HDD stuff\621878F4EG>md5 F4EG.EXE
AD1930D9FE53A6C8E20C3F57936452D1 F4EG.EXE
Click to expand...

What is the HD240UI_JP.EXE file for? I don't see "-JP" on my model number, so I used F4EG.exe. Should I use HD240UI_JP.EXE?
 
mevenson said:
Here's what I get when I do an MD5 on the files I used:

G:\install\Samsung HDD stuff\089312HD240UI_JP>md5 HD240UI_JP.EXE
EEE0492525D836392551021BB744CDCB HD240UI_JP.EXE

G:\install\Samsung HDD stuff\621878F4EG>md5 F4EG.EXE
AD1930D9FE53A6C8E20C3F57936452D1 F4EG.EXE
Click to expand...

I think the first one (_JP) may be a Japanese version of the firmware. I wonder if it was mistakenly posted on the American website, and then taken down? Or did you specifically choose the Japanese support site?
 
That is an update for a different drive (Model : F4EG HD204UI/JP1, HD204UI/JP2).
 
You must log in or register to reply here.
Back
Top