HPE SAS SSD drives fail @ 32,768 hours

Dawg10

Associate
Dec 24, 2016
217
112
43
HPE Support document - HPE Support Center

Bulletin: HPE SAS Solid State Drives - Critical Firmware Upgrade Required for Certain HPE SAS Solid State Drive Models to Prevent Drive Failure at 32,768 Hours of Operation

HPE was notified by a Solid State Drive (SSD) manufacturer of a firmware defect affecting certain SAS SSD models (reference the table below) used in a number of HPE server and storage products (i.e., HPE ProLiant, Synergy, Apollo, JBOD D3xxx, D6xxx, D8xxx, MSA and StoreVirtual 3200 are affected. 3PAR, Nimble and Primera are not affected.)

The issue affects SSDs with an HPE firmware version prior to HPD8 that results in SSD failure at 32,768 hours of operation (i.e., 3 years, 270 days 8 hours). After the SSD failure occurs, neither the SSD nor the data can be recovered. In addition, SSDs which were put into service at the same time will likely fail nearly simultaneously.
 

eddyg

New Member
May 7, 2019
4
5
3
I purchased some used HGST HSCP0480 S5xnNMRI drives from ebay based on a post from @james23.

These drives have HP labels on them, but show up in ssaducli in my HP server as HGST drives (and do not support HP's "Smart Solid State Drive Wearout Status" reporting).

Any idea if these drives are affected? HP's bulletin just says "notified by a Solid State Drive (SSD) manufacturer", but needless to say that HP bulletin is a bit disconcerting!

EDIT: looks like the HGST HSCP0480S5xnNMRI drives were used for HP 3PAR StoreServ 20000 storage servers, and the bulletin says 3PAR is not affected.
 
Last edited:

SRussell

Active Member
Oct 7, 2019
292
147
43
US
If you fail to apply the patch and the drive reaches the failure hour does this mean the drive is 100% unusable?

The way I read the bulletin I assumed once it reaches the failure hour you will be unable to apply the patch, reformat, and continue to use the drive.
 

eddyg

New Member
May 7, 2019
4
5
3
The bulletin says:

After the SSD failure occurs, neither the SSD nor the data can be recovered.

which implies to me the drive itself is toast (e.g. it gets stuck in some loop at boot-up trying to deal with power-on hours that has now gone "negative" because of a signed 16-bit integer or something...)
 

Rand__

Well-Known Member
Mar 6, 2014
4,436
864
113
So the most interesting question is now - is only HPE affected? I looked up some of the HPE SKU's and most seem to be HGST SSDs (if one can trust the pics) (some look like Intel too); but we all know that HP uses their own firmware on those.
Now normally that would imply that only actual HPE drives are affected, but if HPE got notified by the SSD vendor, then the issue is not necessarily in the HPE firmware ( or HPE modifications) since those would not be known to the SSD vendor .

That then means its something the SSD vendor created and *that* might have used in other drives too... (unless HPE outsourced the modification to the SSD vendor which might actually be the case as well. Hm lets hope for that)
 
  • Like
Reactions: Samir

infinitevalence

New Member
Nov 19, 2019
3
1
3
So this lists the MZILS3T8HJCM PM1633 Samsung drives as part of the affected set of drives. Anyone know if this also is a risk with NetApp/IBM, Dell, or direct Samsung branded versions?
 
  • Like
Reactions: Samir

Rand__

Well-Known Member
Mar 6, 2014
4,436
864
113
If you found Samsung and I found HGST drives then the likelihood of being a HPE specific part is high (and vendor related low), so at this point I'd say probably not
 
  • Like
Reactions: Samir

MiniKnight

Well-Known Member
Mar 30, 2012
2,984
888
113
NYC
Are they just Samsung and HGST? I saw some Intel looking SAS drives when I searched but Intel doesn't make SAS drives.

Maybe it'd be good to compile a list.
 
  • Like
Reactions: Samir

BLinux

cat lover server enthusiast
Jul 7, 2016
2,516
958
113
artofserver.com
hmm... it's hard to tell which SSDs these are... image searches are unreliable. I have a bunch of HPE branded HGST HUSMM 400GB and 800GB SSDs... anyone know if these are affected?
 
  • Like
Reactions: T_Minus

Fritz

Well-Known Member
Apr 6, 2015
2,076
456
83
66
Geez, remind me not to buy anymore HP branded SSD's in the future. Why no class action ???????
 

Rand__

Well-Known Member
Mar 6, 2014
4,436
864
113
yes, except i'm not physically near the server that has them. can the HPE p/n be read off with some utility in Linux in a non-HPE server?
Probably not, but you could use smartctl to get the firmware version - if its HPDx (x<8) then the risk is there...

Edit: Actually you might be able to see the "HPE Model Number" in smartctl if they replaced the original drive id string... no idea, dont have any HP SSDs
 
Last edited:
  • Like
Reactions: T_Minus

i386

Well-Known Member
Mar 18, 2016
1,893
492
83
31
Are they just Samsung and HGST? I saw some Intel looking SAS drives when I searched but Intel doesn't make SAS drives.
Intel and hgst developed some controllers together, hgst sold sas ssd and intel sata ssds with these controllers (and Micron/Intel = imft nand :D)
 

Cipher

Member
Aug 8, 2014
147
11
18
49
I purchased some used HGST HSCP0480 S5xnNMRI drives from ebay based on a post from @james23.

These drives have HP labels on them, but show up in ssaducli in my HP server as HGST drives (and do not support HP's "Smart Solid State Drive Wearout Status" reporting).

Any idea if these drives are affected? HP's bulletin just says "notified by a Solid State Drive (SSD) manufacturer", but needless to say that HP bulletin is a bit disconcerting!

EDIT: looks like the HGST HSCP0480S5xnNMRI drives were used for HP 3PAR StoreServ 20000 storage servers, and the bulletin says 3PAR is not affected.
I was wondering the same as my HSSC0400S5xnNMME drives which are from HP, but also show up as HGST in Windows. I read the same 3PAR note so I'm hoping that applies to all drives used in those systems.