Not patched by Oct 2020? Your drives could get bricked…
Various Solid-Point out Drives (SSDs) made by SanDisk experience from a flaw that can see them wiping out every thing saved on them at 40,000 hours (four a long time) — with HPE currently becoming a member of Dell in naming SanDisk proprietor Western Digital as responsible for the bug, which has seen system directors scramble to obtain and fix impacted servers.
Neglecting to get a firmware fix in “will final result in drive failure and information decline at 40,000 hours of procedure and involve restoration of information from backup if there is no fault tolerance, such as RAID or even in a fault tolerance RAID mode if extra SSDs fall short than can be supported by the fault tolerance of the RAID mode on the reasonable drive” HPE said.
It included: “After the SSD failure happens, neither the SSD nor the information can be recovered. In addition, SSDs which have been set into company at the identical time will probable fall short nearly simultaneously.” (Numerous skilled groups will build stacks with non-sequential serial figures and storage items from various distributors, but that’s not usually easy…)
See also: As AWS Slashes Disaster Recovery Costs by eighty%, Can Impartial Corporations Compete?
HPE guidance for consumers posted on March twenty said that primarily based on its examination of when servers equipped with the SanDisk SSDs commenced shipping and delivery, consumers shouldn’t experience issues in advance of Oct 2020 giving close-end users plenty of time to make the critical patch in advance of their drives get bricked. Other OEMs are probable to be impacted.
(Personal computer Small business Assessment has not however seen any further buyer advisories. If you acquired a person from a different server vendor, get in contact with our editor…)
Hey, Western Digital: Many thanks for That
These days naming Western Digital for the 1st time (an before assertion experienced just cited a “Solid Point out Generate company), HPE instructed Personal computer Small business Assessment in an emailed assertion: “HPE was notified by Western Digital of a company firmware defect in particular SAS SSD products utilised across the market.
“Because this defect only will cause drive failure after 40,000 hours of procedure, no HPE consumers are in threat of failing for various months. HPE has obtained Serial Selection information on the drives delivered to HPE consumers, and we are actively achieving out to people consumers and to offer updated firmware.”
Western Digital did not reply to requests for remark. The enterprise before instructed Blocks and Storage that, “Per Western Digital company policy, we are not able to offer comments about other vendors’ items. As this falls inside of HPE’s portfolio, all related solution concerns would greatest be addressed with HPE immediately.” (Which, offered it is in the end a Western Digital solution flaw, seems un-chivalrous…)
SanDisk SSD Bug: Dell Advised Prospects in February
Dell meanwhile notified its consumers in February, emailing them to say that it experienced “identified a likely critical concern exactly where particular good point out drives might working experience failure and prospective information decline due to an concern with the drives’ firmware, the drives might fall short after about 40,000 hours of use.”
SanDisk drives ranging from 200GB to one.6TB are recognized to be impacted. These can be observed in a sprawling array of Dell and HPE servers: both businesses have furnished end users with a comprehensive record of impacted items.
HPE has made Linux, VMware, and Home windows scripts readily available which accomplish an SSD drive firmware look at for the 40,000 electric power-on-hours failure concern, as has Dell, which pointed the finger at SanDisk design figures LT0200MO, LT0400MO, LT0800MO, LT1600MO, LT0200WM, LT0400WM, LT0800WM, LT0800RO and LT1600RO.
Attentive units directors must have minor difficulties determining the servers impacting and patching them in a with any luck , bug-no cost method, but the concern is annoying for significant OEMs like Dell and HPE which face owning to determine and notify all impacted consumers — and which will no doubt acquire the brunt of any criticism.