Samsung 870 EVO - Beware, certain batches prone to failure!

[Daytrader]

What replacement ssd do people recommend, or are 2022 samsungs ok you reckon ? thx

 

[80251]

The root cause of the problem may lie in the drive itself.
The new firmware may appear to fix the problem at first glance, but it may actually be just masking the problem.
We have to note the newly added SMART ID "FC".

What's interesting about your observation is that none of my 850 Pro SSD's (I have three of them) or my 870 QVO have a FC SMART ID statistic

 

[Captain Danger]

I bought 4 pcs of 870 EVO 1 TB in June 2021. The first one failed now with about 3 TB written, can no longer read some of the files on the disk, there are bad blocks and lots of read errors in the smart report. Disk manufacture date is 2021.01.

Power On Hours for the failed drive?

 

[parityerror]

Power On Hours for the failed drive?

9031 hours at the moment. Can't say when the disk failed, it's only one large file that is unreadable at the moment and I hadn't been accessing it recently.

 

[parityerror]

I finally managed to update the firmware on two disks using fumagician. The reason it was crashing before was that when it scanned the disks in the system, it found an SD card reader with no disk in it (I found this out by tracing the program). In the meantime, I also tried Samsung_SSD_DC_Toolkit_for_Linux_V2.1.exe which recognized the disks, but didn't want to update them (saying "[ERROR] The SSD connected does not have a valid update in the given firmware files."). It doesn't officially support consumer drives, but I was desperate enough to try it anyway.
So, I will RMA the failed disk and hope that after the update, the others will work reliably.

After the update, here's some info of the two disks. Whatever the attribute 252 is, the value is strangely high on the less used disk:

ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 0
9 Power_On_Hours 0x0032 097 097 000 Old_age Always - 11292
12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 45
177 Wear_Leveling_Count 0x0013 099 099 000 Pre-fail Always - 1
179 Used_Rsvd_Blk_Cnt_Tot 0x0013 100 100 010 Pre-fail Always - 0
181 Program_Fail_Cnt_Total 0x0032 100 100 010 Old_age Always - 0
182 Erase_Fail_Count_Total 0x0032 100 100 010 Old_age Always - 0
183 Runtime_Bad_Block 0x0013 100 100 010 Pre-fail Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0032 071 045 000 Old_age Always - 29
195 Hardware_ECC_Recovered 0x001a 200 200 000 Old_age Always - 0
199 UDMA_CRC_Error_Count 0x003e 100 100 000 Old_age Always - 0
235 Unknown_Attribute 0x0012 099 099 000 Old_age Always - 28
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 2674519688
252 Unknown_Attribute 0x0032 100 100 000 Old_age Always - 2

ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 0
9 Power_On_Hours 0x0032 099 099 000 Old_age Always - 1468
12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 20
177 Wear_Leveling_Count 0x0013 100 100 000 Pre-fail Always - 0
179 Used_Rsvd_Blk_Cnt_Tot 0x0013 100 100 010 Pre-fail Always - 0
181 Program_Fail_Cnt_Total 0x0032 100 100 010 Old_age Always - 0
182 Erase_Fail_Count_Total 0x0032 100 100 010 Old_age Always - 0
183 Runtime_Bad_Block 0x0013 100 100 010 Pre-fail Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0032 069 056 000 Old_age Always - 31
195 Hardware_ECC_Recovered 0x001a 200 200 000 Old_age Always - 0
199 UDMA_CRC_Error_Count 0x003e 100 100 000 Old_age Always - 0
235 Unknown_Attribute 0x0012 099 099 000 Old_age Always - 4
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 809939957
252 Unknown_Attribute 0x0032 100 100 000 Old_age Always - 22

 

[Daytrader]

So has anyone that have rma there bad 870 ssd, ever got a new one back, or are they just refurbished ssd's ? thx

 

[80251]

@parityerror, the 252 SMART value is the FC value mentioned by E.S.

 

[parityerror]

Thanks, yes, but it seems we don't know for sure what that value actually means (if we should be worried about it increasing or not).

 

[E.S]

I have posted in the past.
People on this forum don't read my posts at all.


SMART ID "FC" (252) is described on wikipedia as follows:

252 0xFC

Newly Added Bad Flash Block

The Newly Added Bad Flash Block attribute indicates the total number of bad flash blocks the drive detected since it was first initialized in manufacturing.[81]

 

[parityerror]

I have posted in the past.
People on this forum don't read my posts at all.


SMART ID "FC" (252) is described on wikipedia as follows:

252 0xFC

Newly Added Bad Flash Block

The Newly Added Bad Flash Block attribute indicates the total number of bad flash blocks the drive detected since it was first initialized in manufacturing.[81]

I have read the whole thread including your messages, but there was some disagreement about the meaning of this smart parameter..

 

[80251]

Is Samsung still manufacturing 870 EVO's? Or have the moved on to 970 EVO's? This whole thread has made me paranoid about my 870 QVO 4 TiB.

 

[E.S]

Thanks, yes, but it seems we don't know for sure what that value actually means (if we should be worried about it increasing or not).

With all these conditions, why do you think there is no need to worry?
That's a little unusual.

1, SMART ID "FC" is an item related to this issue because it was added by updating the firmware.
2, According to wikipedia, SMART ID "FC" is explained as Newly Added Bad Flash Block, and its value continues to increase during use.
3, Samsung doesn't explain anything. (If the problem is solved, you should say it's solved)


As far as I read on wikipedia, I think that SMART ID "FC" (252) is originally an item representing the initial bad block.

What worries me is the possibility of trying to cover up the problem by counting bad blocks acquired during use as early bad blocks.

 

[Daytrader]

So has anyone that have rma there bad 870 ssd, ever got a new one back, or are they just refurbished ssd's ? thx

Anyone ?

 

[ThrashZone]

Hi,
If they send back a refurbished it will state it on the ssd label.
Crucial does this not sure what sammy does though.

Best to contact sammy directly.

 

[Daytrader]

Hi,
If they send back a refurbished it will state it on the ssd label.
Crucial does this not sure what sammy does though.

Best to contact sammy directly.

Ok thx for reply.

 

[Mussels]

Anyone ?

If you ask and no one answers, that means no


Theres maybe 20 people here of the 8 billion in the world, it's not somewhere you're gunna get random people answer that question

 

[ThrashZone]

Hi,
lol
No, it's not a new one sent back using rma
No, it's not a refurbished one sent back using rma
Did I get that right if nobody responds

 

[Daytrader]

Hi,
lol
No, it's not a new one sent back using rma
No, it's not a refurbished one sent back using rma
Did I get that right if nobody responds

Thx for reply

 

[ThrashZone]

Thx for reply

Hi,
Thankfully I've not had to rma any sammy ssd's yet
Pretty sure it can likely be a refurbished and what ever warranty time is left is continued as it started with the original

Not to say refurbished is bad I still have a crucial still kicking 4 years after getting a refurbished replacement linux killed so don't let that detail get you down.

But that is a good question

 

[Daytrader]

I asked because, how would they even be able to refurbish a ssd in the first place, its not mechanical so no moving parts to go wrong, so they should only offer brand new, but i have read other places they did get a refurb of the same year ssd that are the problem.

 

[ThrashZone]

I asked because, how would they even be able to refurbish a ssd in the first place, its not mechanical so no moving parts to go wrong, so they should only offer brand new, but i have read other places they did get a refurb of the same year ssd that are the problem.

Hi,
I'd say that would be the exception if it's hardware failure
In my crucial case it was a firmware issue so they just cleaned and reloaded the firmware
Sammy's to I've read they sent out firmware updates to solve the issue, not sure if it does or not, not many have returned with update details sadly.

 

[R0Sch]

Another one bites the dust!
SSD 870 EVO 1TB, made in China date 05.2021, firmware SVT01B6Q used exactly 1 year with only 4.5 TBW is having 265 bad blocks, read/write errors (can't fully backup data or install new games any more) and after each scan increasing ECC Error Rate / Uncrorrectable Error Count. Crystal Disk Info shows it's health at 77% and Disk Sentinel already at 57%. Sent the info's to Samsung for RMA start.
Here are the S.M.A.R.T data before/after Full Diagnostic Scan and the full scan results

I'm really worried about having the same drive and capacity in another device where it's an OS main drive. That one is from 04.2021, but the S.M.A.R.T values are ok since I only used it for 280h and 2.3 TBW.

 

[Mussels]

Another one bites the dust!
SSD 870 EVO 1TB, made in China date 05.2021, firmware SVT01B6Q used exactly 1 year with only 4.5 TBW is having 265 bad blocks, read/write errors (can't fully backup data or install new games any more) and after each scan increasing ECC Error Rate / Uncrorrectable Error Count. Crystal Disk Info shows it's health at 77% and Disk Sentinel already at 57%. Sent the info's to Samsung for RMA start.
Here are the S.M.A.R.T data before/after Full Diagnostic Scan and the full scan results

I'm really worried about having the same drive and capacity in another device where it's an OS main drive. That one is from 04.2021, but the S.M.A.R.T values are ok since I only used it for 280h and 2.3 TBW.

Can you confirm TRIM is running on your OS?

In a cmd prompt:

fsutil behavior query DisableDeleteNotify

If you get a result of zero (0), then it means that TRIM is enabled

If it's a 1, then run

fsutil behavior set DisableDeleteNotify 0

It's about the only thing i can think of that could accelerate such an issue

 

[R0Sch]

Can you confirm TRIM is running on your OS?

It's about the only thing i can think of that could accelerate such an issue

Yes, I can confirm TRIM is enabled. It's not the only SSD in my PC. The other ones (Crucial M4 and C300) are much older are still showing 0 bad blocks and 0 reallocated sectors after 25000 hours of usage. I use my SSD's only for games and downloads.

The only thing I could think of is maybe there's a weird incompatibility with the Marvell SATA III controller (X58 Sabertooth) and Samsung SSD's in particular. My 870 Evo was shown as "this drive is not supported" in Samsung Magician until I switched it to a Sata II port, just to be able to get the SMART and diagnostic results for their RMA. My M4 did not develop such issues albeit being connected to the other SATA III port.

Wonder if I should format the drive and update to the latest firmware SVT02B6Q and keep using it as another user suggested. The damage is there so it would only mask the problem and probably reduce the drives total capacity. The bad blocks would be excluded from usable space, right?

 

[Mussels]

Googling says x58 supports TRIM just fine, except it wont work in RAID

Was that extra controller in a RAID mode?