Seagate SNAFU sees Cisco servers primed for data loss
Disks shipped with the wrong write cache settings and found their way into UCS boxen
Cisco's dropped Seagate in the pooh for a mess that's seen some UCS servers released into the wild in configurations susceptible to data loss.
The mistake was small and simple: as Cisco's Field Notice explains, “Cisco ships all of their hard drives from manufacturing with drive write cache disabled” because “If a drive on one of the affected platforms has drive write cache enabled, and the cache is not flushed prior to power loss, data that still resides in the drive cache will be lost.”
Deep in the Field Notice is the name Cisco's applied to it for its internal equivalent of a CVE: “Seagate MakaraBP1/2/4TB SAS Write Cache Mode Enable when delivering to Cisco”.
Which rather sounds like Seagate sent Cisco drives with the cache enabled when Cisco prefers it to be well and truly off. And that Cisco didn't check all the disks its received for a while, as the Notice also says "During a quality audit, select units were found to have the drive write cache enabled."
The good news is that with a little tweaking, you can turn the cache off. The bad is that one of the three UCS variants impacted is the UCSC-C3X60, a "dense storage rack server, optimized for large datasets” and which can house up to 56 disks. So there may be rather a lot of tweaking to be done. ?