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

[999cgm]

An update for my Samsung SSD 870 EVO 500GB , Looks like the 252 (0xfc) incremented a little more.. again I am on linux. It seems smartmontools (version 7.5) now have a name for this 252..

smartctl 7.5 2025-04-30 r5714 [x86_64-linux-6.12.27-1-lts] (local build)
Copyright (C) 2002-25, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family: Samsung based SSDs
Device Model: Samsung SSD 870 EVO 500GB
Serial Number: S7EWNJ0X102368Y
LU WWN Device Id: 5 002538 f3412275e
Firmware Version: SVT02B6Q
User Capacity: 500,107,862,016 bytes [500 GB]
Sector Size: 512 bytes logical/physical
Rotation Rate: Solid State Device
Form Factor: 2.5 inches
TRIM Command: Available, deterministic, zeroed
Device is: In smartctl database 7.5/5706
ATA Version is: ACS-4 T13/BSR INCITS 529 revision 5
SATA Version is: SATA 3.3, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is: Mon May 12 10:39:21 2025 EEST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled



Vendor Specific SMART Attributes with Thresholds:
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 - 3095
12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 927
177 Wear_Leveling_Count 0x0013 099 099 000 Pre-fail Always - 2
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 Uncorrectable_Error_Cnt 0x0032 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0032 072 050 000 Old_age Always - 28
195 ECC_Error_Rate 0x001a 200 200 000 Old_age Always - 0
199 CRC_Error_Count 0x003e 099 099 000 Old_age Always - 1
235 POR_Recovery_Count 0x0012 099 099 000 Old_age Always - 17
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 3127924722
252 Added_Bad_Flash_Blk_Ct 0x0032 100 100 000 Old_age Always - 4

 

[SuperConker]

Well yes these blocks are lost. But that is not a big issue as, just like you've said, SSD drives have quite a bit of spare area. Normally these spare blocks are used transparently - many drives have a "SSD life left" metric for that. In case of Samsung they expose a "wear levelling count" metric and a "used reserve blocks count". Over time these counters in Samsung SSDs will increase

So it's perfectly normal for any SSD drive to remap bad blocks. That's part of the wear leveling algorithms in their controller.

In our particular case of the early 870 Evo firmware, the controller failed detect to dying blocks so it could not reassign them proactively/transparently. These blocks now show up as "reallocated sectors count". Now that has caused actual issues for people (a.k.a. data loss in these blocks). But that is/was an issue within the controller firmware which can and has been fixed by an update.

And BTW: my drive is still in warranty and Samsung denies to swap it because it's not defective. It passes all tests and the bad blocks count did not increase anymore after the FW update - I have written many terrabytes to that drive ever since.

It is not a good idea to disregard bad blocks/sectors on an SSD (especially not on the Samsung 870 EVO).

It is also not normal for an SSD to develop bad blocks within the rated TBW of the drive.

If your drive came with SVT01B6Q or SVT02B6Q Firmware then it's the older version of the drive
(which cannot be updated to the latest SVT03B6Q Firmware).

And if you have any important data on it you will eventually lose access to that data
(and it will not be possible to backup/recover that data, even if you can still see it).

 

[Hardcore Games]

One take away I see is that the 870 EVO seems to not have much of any spare blocks to swap out to the good pool. This suggests the need to use CHKDSK /R occasionally to sweep the nblocks ands remove bad ones manually,

 

[PowerBuilder]

The real reality is, don't use TLC (3bit MLC). And Samsung SSDs have too small SLC mode sizes. There are almost no 2bit MLC sections, and when you fill up the capacity, it always operates as TLC, which is when planned obsolescence begins. When Samsung 840 EVO 120GB (Samsung's first TLC SSD at the time) first came out, I trusted it and bought it, but it made me angry because of its unsmooth operation (I am very sensitive to even small differences). So I tore off the aluminum SSD housing and exposed the internal circuit board of the 840 EVO like an NVME card (it's like peeling off an animal skin by removing the metal shell of a 2.5-inch SSD), but the circuit board itself was left intact, and I gave it to someone else's old computer (with a slow HDD) that I used lightly for a free upgrade, and I was dramatically freed from the clutches of TLC SSDs. The old days when fixed 2bit MLC SSDs were mainstream were better. The now discontinued Micron MX200 250GB operated in SLC mode when filled less than halfway. Never use TLC, QLC, etc. unless you are going to use them as ROM for data backup (read-only mode). The unethical commercial practices of SSD companies are clear. In other words, they gave up the amazing 50,000 to 100,000 write lifespan of SLC SSDs, higher write (read) performance, stability, and very fast response speed, and started with 2-bit MLC, then 3-bit MLC (TLC), 4-bit MLC (QLC), inflated the capacity by 2-4 times and made it thinner (the truth is that the capacity increases by 1 as you go from SLC->MLC->TLC->QLC, but the performance, stability, and especially the write lifespan are reduced by 10 times). This trick of selling them by inflating the capacity like an easily popped balloon and planned obsolescence is the sad reality of most SSDs being sold now. MTRON SSD, which was the starting point of SSD popularization, was expensive but was SLC (My first SSD was MTRON 64GB and the performance was around 100MB/S. This old MTRON SLC SSD will be more stable than the latest TLC SSD.) I hope Micron RealSSD P300 SLC SSD and Intel X25-E SLC SSD will be revived. Or they will revive Optane memory.

 

[Jon555]

It is not a good idea to disregard bad blocks/sectors on an SSD (especially not on the Samsung 870 EVO).

It is also not normal for an SSD to develop bad blocks within the rated TBW of the drive.

If your drive came with SVT01B6Q or SVT02B6Q Firmware then it's the older version of the drive
(which cannot be updated to the latest SVT03B6Q Firmware).

And if you have any important data on it you will eventually lose access to that data
(and it will not be possible to backup/recover that data, even if you can still see it).

I have two 870 2TB drives, my older one is from June 2022 with firmware SVT02B6Q and one from Sept 2024 with firmware SVT03B6Q. I presume the different firmware relates to the change from 128 layer (aka V6) to 133 layer (aka V6 Prime) flash in 2023.

I specifically asked Samsung support about the issue mentioned in this thread in December 2023 and they said my 2022 drive had firmware that post-dated the issue they had, in not the greatest English:..

"This is the firmware related problem and was improved at the middle of 2021-December already.
The update of the improved firmware was released from the December of 2021 and the products with the latest firmware don't have this problem."

The 252 Raw values are 0 for the new one (just less than one TB written) and D4 for the old one (11.5TB written). The processed values are 100 for each and the minimum ever is also 100, the limit is 0.
I'm not in the slightest bit concerned...

P.S. The 870 EVO are tlc (3 bits per cell) which Samsung unhelpfully call 3-bit mlc (where mlc is pretty universally used for 2 bits per cell). You can use tlc cells as slc for caching though (I think almost no-one puts separate slc for cache these days, there is also 2GB of DRAM)... While tlc has a lower write life than (real) mlc they are both long enough not to be an issue (my 2TB drives are rated for 1,200 TB of writes by Samsung). BTW The 870 QVO is qlc - 4 bits per cell (i.e. 16 levels).

 

[SuperConker]

The update of the improved firmware was released from the December of 2021 and the products with the latest firmware don't have this problem."

Many of the 2022 drives (and even some 2023 drives) with the SVT02B6Q Firmware on them have been failing too.

 

[Jon555]

Many of the 2022 drives (and even some 2023 drives) with the SVT02B6Q Firmware on them have been failing too.

What is the evidence for this please?

 

[999cgm]

for my 870 EVO ., is strange that I have 4 Bad blocks(252) and 179 Used_Rsvd_Blk_Cnt_Tot is 0 , and , 5 Reallocated_Sector_Ct is 0

 

[999cgm]

Any theory about what I see with my SSD? what is the ideea behind Bad blocks(252/0xfc) that is slowly incremeting? BTW for me is still 4 , I just do not understand when is incrementing..
if is not doing realloc (Reallocated_Sector_Ct is 0 and Used_Rsvd_Blk_Cnt_Tot is also 0 )

 

[CiTay]

The main SMART attributes to check for the issue covered in this thread are "Reallocated Sector Count", "Used Reserve Block", "Runtime Bad Block", as well as "Uncorrectable Error Count" and "ECC Error Rate".

The SMART ID "FC" is - at least in CrystalDiskInfo - only named "Vendor Specific", and on my unaffected 870 EVO 4TB, range between 800-1600, and also in that region on my previously affected drive (which, since the initial fix and firmware update, now behaves normally again for quite some years, with no further issues).

So i believe what you are seeing is of no relevance, as long as none of the critical attributes listed in my first paragraph give any cause for concern by not being zero anymore.

 

[999cgm]

OK. Thank you , I can confirm in my case all those SMART counters are on zero.. here is the current state:

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       -       3228
 12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       964
177 Wear_Leveling_Count     0x0013   099   099   000    Pre-fail  Always       -       2
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 Uncorrectable_Error_Cnt 0x0032   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0032   073   050   000    Old_age   Always       -       27
195 ECC_Error_Rate          0x001a   200   200   000    Old_age   Always       -       0
199 CRC_Error_Count         0x003e   099   099   000    Old_age   Always       -       1
235 POR_Recovery_Count      0x0012   099   099   000    Old_age   Always       -       17
241 Total_LBAs_Written      0x0032   099   099   000    Old_age   Always       -       3217229154
252 Added_Bad_Flash_Blk_Ct  0x0032   100   100   000    Old_age   Always       -       4

For ref, the commit where this 252/0xfc was named Added_Bad_Flash_Blk_Ct in smartmontools:

drivedb.h: · smartmontools/smartmontools@bc07dc0

- Apacer SSDs: AP1TPPSS25-R (#1640). - Crucial/Micron RealSSD C300/P300: C300-MTFDBAK* (#1421), MTFDBAK* (#1728). - Micron 5100 / 52x0 / 5300 / 5400 SSDs: 5200 240GB. - Intel 53x and Pro 1500/250...

github.com

 

[CiTay]

Yep, all looks pretty normal, here's from my two drives who never had any sign of problems:

If you're wondering about the very low drive temperatures everywhere, i just booted up two minutes ago.

And here's the previously affected drive, but running strong ever since (no problems and none of the critical values have increased):

 

[999cgm]

Thank you very much @CiTay for detalied info !!

 

[proedge]

Hey guys, I'm having a little trouble with a newly purchased 870 EVO. The box says it's manufactured date 2025 04 20, and the serial number ALMOST matches the one that shows up in CrystaldiskInfo... the one that shows up in the software has an A on the end. Not sure if that's normal? When I load up Samsung Magician, it does not recognize the drive and gives me an "Invalid serial number" error. The firmware is SVT03B6Q... everything else seems to work normally. I was able to clone my old SSD onto this new one without any issues (I used miniTool partition wizard). Just concerned that this drive might be a fake, as I got it from Amazon.
Any ideas?

 

[CiTay]

Well, you should contact Samsung, give them all the information, they can tell you if anything is shady about it.

 

[Aiko]

Hello guys, I need a help about the SSD. I do not real all 51 pages unfortunately only 2 3 pages.

I have this SSD and lost my important data's for work. How can recover the data's from this SSD?

 

[CiTay]

Well, how is the state of the SSD? Normally, only a fraction of files are affected once this issue has appeared. Most of the files, you can just copy to a different drive without issue. On some files, it will get stuck, because it cannot read parts of the file properly anymore. Those files can, if anything, only be partly recovered, they may also be a total loss (if you don't have a backup). Use FreeFileSync for copying the data, it will show a log afterwards of which files couldn't be copied.

Of course, doing the firmware update is also important. This way, i could make my failed drive fully usable again for later. I'm still using it as we speak.