Samsung PM1643 2019.09 date code - 32k hour bug

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

yukas

Member
Jun 3, 2022
36
2
8
Some forums say that such disks can still be brought back to life by resetting SMART - has anyone succeeded?
I have several such disks and I'm wondering how to flash/reset them in HBA MODE.
 

Tickety-boo

Member
Sep 8, 2019
47
5
8
Nope, this error message states the SSD died as the bug (timer expired after 32k hours runtime) happened. As far as I know nobody was ever able to re-activate any disk which died that way.
Sorry, but it’s gone forever.
The OEM solution is to update the FW version, doesn't that save the 32K hours bug?
 

yukas

Member
Jun 3, 2022
36
2
8
bump
anybody has been reflash sas ssd pm1643a with same problems?
Samsung PM1643 2019.09 date code - 32k hour bug
 

krakonos

New Member
Oct 21, 2023
21
5
3
Hi!

I'm looking at some MZILS3T8HCJM in revision CQP2. Seems it might be OK, as I've seen one with 56k hours. I wonder if somebody can confirm that? I can't find reference to the CQP2 firmware anywhere and if the last letter is revision in order, it might be an earlier revision (not that it indicates something).

I'd be happier if somebody could confirm this. Also a thread with different issues and revisions would be nice to have, there seems to be a lot of weird bugs in the generation easily available on the 2nd hand market.

Thanks!
 

Ciciban

New Member
Feb 5, 2025
11
0
1
Yes, it's a pain. They count firmware for enterprise stuff to 'service' they want you to pay. Maybe legit for the time before EOL, but after that...come on, just dump all the files in a folder on the ftp-server. This saves headaches and the environment. :)

Meanwhile some info for PM1633, that should be DELL:

MZILS1T9HCHP | MZILS3T8HCJM | MZILS480HCGR | MZILS960HCHP sentenced to death on CQPG, CQPH -> fixed on CQPJ
MZILS15THMLS0D4 | MZILS7T6HMLS0D4 sentenced to death CQN1 -> fixed on CQN3
MZILS1T9HEJH0D4 | MZILS3T8HMLH0D4 | MZILS480HEGR0D4 | MZILS480HEGR0D4 sentenced to death on CQL1, CQL3 -> fixed on CQL5

Edit:
Attached files, ripped out from scos-official-R07.01.04.004.03.zip
I think they should work as .bin or .lod when extracted
What about PM1633a, they have the same problem ?
P/N: MZILS3T8HMLH - 000C3
Product ID: PA33N3T8 EMC3840
Revision: EQL8
 

mr44er

Active Member
Feb 22, 2020
168
51
28

ogplayer

New Member
Aug 15, 2024
20
0
1
This is interesting. The cqpj firmware appears to be incompatible with MZILS3T8HCJM drive. Either that or the firmware needs to be extracted a different way.
 

Attachments

Ciciban

New Member
Feb 5, 2025
11
0
1
I don't know, don't have them. Just have the firmware, dig through this:
Thank you.

But it seems that there is no FW for PM1633A.
 
Last edited:

steve1

New Member
Feb 14, 2025
4
2
3
Question: Does it matter if you use 'generic' firmware, such as the ones from HDD Guru on vendor drives? For example I have some Dell PM1643 and I'll use the drivers from earlier in the thread but wondering what the difference in firmware is between the generic stuff and the vendor provided firmware?
 

mr44er

Active Member
Feb 22, 2020
168
51
28
It won't work, the drive rejects foreign firmware. Foreign in the sense of 3rd party vendor.