In a second SSD snafu in as a few years, Dell and HPE have revealed that the 2 distributors have shipped enterprise drives with a crucial firmware bug, one will ultimately trigger knowledge loss. The bug, seemingly associated to an inner runtime counter within the SSDs, causes them to fail as soon as they attain 40,000 hours runtime, shedding all knowledge within the course of. As a consequence, each corporations have wanted to problem firmware updates for his or her respective drives, as prospects who’ve been operating them 24/7 (or almost as a lot) are beginning to set off the bug.
Ultimately, each points, whereas introduced/documented individually, appear to stem from the identical fundamental flaw. HPE and Dell each used the identical upstream provider (believed to be SanDisk) for SSD controllers and firmware for sure, now-legacy, SSDs that the 2 pc makers bought. And with the oldest of those drives having reached 40,000 hours runtime (four years, 206 days, and 16 hours), this has led to the invention of the firmware bug and the necessity to rapidly patch it. To that finish, each corporations have begun rolling out firmware
As reported by Blocks & Files, the precise firmware bug appears to be a comparatively easy off-by-one error that none the much less has a major repercussion to it.
The fault mounted by the Dell EMC firmware issues an Assert operate which had a nasty test to validate the worth of a round buffer’s index worth. Instead of checking the utmost worth as N, it checked for N-1. The repair corrects the assert test to make use of the utmost worth as N.
Overall, Dell EMC shipped a lot of the defective SAS-12Gbps enterprise drives over time, ranging in capability from 200 GB to 1.6 TB. All of which would require the brand new D417 firmware replace to keep away from an premature dying at 40,000 hours.
Meanwhile, HPE shipped 800 GB and 1.6 TB drives utilizing the defective firmware. These drives had been, in flip, had been utilized in quite a few server and storage merchandise, together with HPE ProLiant, Synergy, Apollo 4200, Synergy Storage Modules, D3000 Storage Enclosure, and StoreEasy 1000 Storage, and require HPE’s firmware replace to safe their stability.
As for the provider of the defective SSDs, whereas HPE declined to call its vendor, Dell EMC did reveal that the affected drives had been made by SanDisk (now part of Western Digital). Furthermore, primarily based on a picture of HPE’s MO1600JVYPR SSDs revealed by Blocks & Files, it might seem that HPE’s drives had been additionally made by SanDisk. To that finish, it’s extremely doubtless that the affected Dell EMC and HPE SSDs are basically the identical drives from the identical maker.
Overall, that is the second time in lower than a 12 months {that a} main SSD runtime bug has been revealed. Late final 12 months HPE bumped into the same problem at 32,768 hours with a unique collection of drives. So as SSDs at the moment are dependable sufficient to be put into service for a number of years, we will begin seeing the long-term impression of such an extended service life.
Sources: Blocks & Files, ZDNet
