PSA: HPE Solid State Drive Models, Drive Failure at 32,768 Hours of Operation

[garm]

SUPPORT COMMUNICATION - CUSTOMER BULLETIN
Document ID: a00092491en_us
Version: 1
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

Document Display | HPE Support Center

support.hpe.com

 

[garm]

This seams to be the root cause of some recent spectacular catastrophes in Swedish healthcare and law enforcement...
Redundancy is no solution when all drives fail at the same instance..

 

[cobrakiller58]

That's a perfect example of a reason to have the drives in your pools different ages..... If you had a realtime backup on an HDD you would be fine but who does that of a SSD pool? Think there are going to be law suits about this?

 

[garm]

Sweden doesn't have anything like English common law, so the ability sue for damages is limited

 

[cobrakiller58]

Also interesting is some of these are the massive samsung SSD's we were talking about in another thread in regards to the death of HDD's....... I wonder if these will start appearing cheaply on ebay....

 

[garm]

Apparently the HPE SSD firmwares where 32 bits (27 effective?) instead of 64, thus the register ran out after about 3 years. Hopefully some manager somewhere is fired at least..

 

[Arwen]

Another way to plan around these types of errors, is mixed models or even vendors.

For my FreeNAS pool, I used half WD Reds and half WD Red Pros. I hope that they will have different life spans. (And they were all bought at different times.)

Some people think that is over-kill, but I wanted something to last 5 years or more. (Of course, it was the IPMI firmware & Intel CPU issue that will actually kill my FreeNAS.)