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

[massaker]

Is this temporary?

Just 1 minute later in the video he explains it. Yes, it's temporary, ~5 minutes later it regenerates - but only 4GB it guaranteed, even at fuller state. So the size of pSLC is dynamic between 22 (empty state) and 4GB (almost full drive).

 

[SPDIF]

Why would you worry about that? Will you write 22GB per day? And the top speed of SATA is 550MB/per second max. It's just a buffer to speed some things up, but it stays SATA. If you want faster speeds you should upgrade to NVMe SSD. Then you get speeds of 7000MB/s and even higher.

 

[tekbro]

Thank you all!

 

[Spirch]

opinion on this?

been using it as nvr drive 24/7 recording for 4 ip camera

 

[AusWolf]

View attachment 353784

opinion on this?

been using it as nvr drive 24/7 recording for 4 ip camera

Still 99% left after 85 TBW is really good! Or you have some strange life indicator on that drive. Anyway, if you're not noticing any errors, I don't think you have anything to be concerned about.

 

[massaker]

opinion on this?

How old is the SSD? Used with old FW first? Unfortunately, you blended out the S/N, so i can't see the month of manufacture.

 

[jsfitz54]

you blended out the S/N, so i can't see the month of manufacture.

You don't need to be slumming for someones serial number.

 

[massaker]

You don't need to be slumming for someones serial number

Huh, can you give one reason why someone would want to have someone else's serial number?

 

[Wirko]

Still 99% left after 85 TBW is really good! Or you have some strange life indicator on that drive. Anyway, if you're not noticing any errors, I don't think you have anything to be concerned about.

It could be that the NVR is very easy on the SSD, and does enough write buffering and caching, so the writing is less random and in large enough blocks, with low write amplification as a result.

 

[Spirch]

How old is the SSD? Used with old FW first? Unfortunately, you blended out the S/N, so i can't see the month of manufacture.

based on the serial number, I see W7 so July 2023?

I got it last September

It could be that the NVR is very easy on the SSD, and does enough write buffering and caching, so the writing is less random and in large enough blocks, with low write amplification as a result.

I use the term nvr but it's inside a minipc (EQ12 from beelink - windows 10) using samba share

 

[massaker]

W7 so July 2023?

Yes, W = 2023

got it last September

Absolutely impossible. Simple maths:
As of yesterday: you're using it since 308 days, 02. Juli 2024 minus 308 days = even with 24/7 use you're using it since 29. Aug 2023.

 

[CiTay]

opinion on this?

Textbook unaffected 870 EVO, couldn't be any better.

The markers i mentioned in my first post are still the ones to look for:
Elevated "Reallocated Sector Count", "Used Reserve Block" and "Runtime Bad Block" count - this would be the first warning sign
Non-zero "Uncorrectable Error Count" and "ECC Error Rate", and especially if those two keep rising when you read/write files - this would be the definitive sign

You have those all on zero, so it's all good.

 

[Spirch]

Yes, W = 2023

Absolutely impossible. Simple maths:
As of yesterday: you're using it since 308 days, 02. Juli 2024 minus 308 days = even with 24/7 use you're using it since 29. Aug 2023.

I was close, but you are closer

my bad :-(

 

[xorbe]

I am still not clear -- if the firmware is upgrade, is an 870 EVO still a time bomb?

 

[CiTay]

With the firmware updated and the bad sectors remapped, even my affected drive became completely usable again. No trouble from my other two 870 EVO.

 

[Sora]

With the firmware updated and the bad sectors remapped, even my affected drive became completely usable again. No trouble from my other two 870 EVO.

there is not and has never been a firmware fix for 870 evo's, the failure was contaminated nand.

 

[CiTay]

Well, i had no further problems with that drive since 2 1/2 years ago, and i didn't spare that drive at all, i purposely used it hard. So, make of that what you will.

 

[burnerdosrania]

Hi, is updating firmware from SVT01B6Q to SVT02B6Q stopping the drive errors from further increasing on already affected SSD? I really can't send it to RMA cuz it's my main drive, i'm using it on a daily basis.
Yesterday I noticed some errors in windows explorer while copying files from my internal 1TB 870 to 500GB 860 connected via SATA->USB adapter.

I've checked SMART stats in Magician and there are 99 Uncorrectable Error Counts and 199 ECC Error Rate. I've immediately updated firmware when I saw it.

 

[Zeromis]

Hi, is updating firmware from SVT01B6Q to SVT02B6Q stopping the drive errors from further increasing on already affected SSD? I really can't send it to RMA cuz it's my main drive, i'm using it on a daily basis.
Yesterday I noticed some errors in windows explorer while copying files from my internal 1TB 870 to 500GB 860 connected via SATA->USB adapter.

I've checked SMART stats in Magician and there are 99 Uncorrectable Error Counts and 199 ECC Error Rate. I've immediately updated firmware when I saw it.

Updating the firmware should stop any further issues yes. That said id really recommend at the very least that you copy all the files off it and do a full format(not a short one)/zero the drive or something so the drive sorts out any bad sectors. Until you do that you might still come across lingering issues.

I've been using an affected drive with the new firmware for around two years now and I've had no new issues crop up.

 

[Boblob]

Hello, is it possible that a similar problem appears on ssd 860 EVO?

I bought in December 2018 two 860evo (250 and 500gb). When one of the two disks is used for windows, it has happened to me (particularly several months ago and again yesterday) that the disk displays 100% use and causes the pc to freeze.

I only solved this freeze problem (which happened with both hard disks and windows installed on them) by installing windows on another disk (980 pro).

I've attached the CrystalDisk images for the two hard disks I'm talking about.

The amazing thing is that I haven't had any data loss problems when using the 500gb as storage.
Do you think the problem lies with the hard drives?
Maybe Windows with its specific hard drives?


Thanks for you time !

 

[CiTay]

Can you set it to English though?

The 860 EVO is not affected by this issue, and you have no telltale signs of anything like what can happen to some 870 EVO. What strikes me as odd though, you have high "CRC Error Count" (ID C7). This is not that common, this can sometimes happen with bad SATA cables for example, or maybe an otherwise slightly unstable system. Another thing, you have high "POR Recovery Count" (ID EB), this value means, how often the SSD suddenly lost power (for example from a reset on the PC, from the PC shutting down unexpectedly).

As far as actual errors in the flash cells, you have zero on both drives. So this is not about what is covered in this thread whatsoever. You should perhaps make a new thread, where you list all your hardware including the PSU model and everything, maybe you just have a general problem with your PC. But the SSDs look ok, apart from those two weirdly high values (but they don't point to a problem with the NAND flash memory).

 

[Boblob]

I thought I'd finally found the exact reason for my problem, but it seems not.

Thanks for your reply, it's already given me some additional insight.
I'll probably try changing the SATA cables and keep a close eye on these hard drives.

 

[SPDIF]

Both of your drives are 100 percent healthy, i would not worry about that.

You have a high CRC error count with means a bad Cable or bad connection on the mobo. Also a bad driver can cause this error to arise. It's a pure hardware error and you have to do some research to find the fault. Also a wrong setting in the bios could cause this fault. Try a new cable and another SATA slot, look in the BIOS it is set to AHCI and not RAID.

You also could try them in another PC to see what happens, but i am sure the drives are okay. If you see the same problem with all SATA ports and it works good on another Computer, i could be very well that your mobo is bad. Also try to install the latest drivers from AMD or Intel.

Edit; I see you made another topic about this, well i did answer here but it's the same.

 

[RJARRRPCGP]

Try a new cable and another SATA slot, look in the BIOS it is set to AHCI and not RAID.

Fun fact: Especially on some legacy-BIOS-era Dell PCs, "RAID" is the same thing as AHCI! I came across this on a socket 775 system. If you wanted AHCI, you select "RAID" in the BIOS, LOL.

 

[SPDIF]

Fun fact: Especially on some legacy-BIOS-era Dell PCs, "RAID" is the same thing as AHCI! I came across this on a socket 775 system. If you wanted AHCI, you select "RAID" in the BIOS, LOL.

Yes to make it easier for people... I do remember that from long time ago. And yes now it's NOT Raid...

I also have an old 860EVO here from 256GB that still works okay, Zero CRC errors, but good used already.