Hewlett Packard Enterprise has issued a notice about some of its solid-state hard drives: they have a defect that causes the crash of the drive after exactly 32.768 hours of operation (3 years, 270 days and 8 hours).



A firmware's bug?

It is possible: the number is a power of 2 (215), so it is probably related to an hour counter stored in a variable (a short int?) that reaches its limit.
Anyway, if the firmware is not updated, the disk will lose all data and will no longer be usable.

The bulletin [1] includes links to updates, which at the moment are not available for all defective models: the bug affects 20 SSD model numbers, and to date, HPE has only patched eight of them.

The remaining 12 won’t get patched until the week beginning 9 December 2019.


Is my SSD going to die?

You can follow these HPE instructions [2] to find out how many hours of operation a SSD has collected:

In Linux and Windows, use the ‘SSA’ command to generate ADU reports:
     Use ‘ssa –diag –f ’ to generate a full Diagnostics Report .
     Run as Administrator/root
 For VMware from a local host, execute the following commands to show detailed physical disk information:
     ESXi 5.5 -> /opt/hp/hpssacli/bin/hpssacli ctrl slot=0 pd all show detail
     ESXi 6.5 -> /opt/smartstorageadmin/ssacli/bin/ssacli ctrl slot=0 pd all show detail
 To capture ADU in Vmware, SSA CLI must be installed on a VSphere client:
     Use the ‘ssaduesxi’ command from the client to generate wear gauge and ADU reports.
     Must specify sever to capture data from, user, password etc.
     From a client with vSphere CLI 6.5 and HPE SSA 3.10.3.0, execute the following command to collect the ADU report:
         ssaduesxi --server= --user= --password= --thumbprint= --file=.zip
 Power On Hours can also be seen on the SSD Wear Gauge section for each drive in the ADU report:


References

  1. HPE SAS Solid State Drives - Critical Firmware Upgrade Required...
  2. Smart Storage Administrator (SSA) – Quick Guide to Determine SSD Power on Hours