[ajross]

It's not about trusting the storage, it's that failures in components not storage end up junking the storage in the repair. In all likelihood this board suffered some other sort of component failure (because if it was the flash, something would still boot and display). That 1TB flash device was perfectly good, and with only an ever so slightly compromised design could have been dropped right into a new board and the user would have left happy with "their laptop" still intact.

But now the poor guy needs to recover a huge partition from partial backups, and that sucks. And it was totally avoidable..

(FWIW: I don't understand your point about not trusting storage on laptops. Modern SSDs are extremely reliable. Sure they can fail, and you should always have a plan for what happens if they do. But in practice you should absolutely trust your laptop storage.)

[ska]

> and you should always have a plan for what happens if they do.

This is what I meant. Not that you expect your laptop to fail (or be stolen), but any important work should exist in a recoverable way. OP claims to have lost a lot of data when this happened to them. That should approximately never happen, at least not more than the day's worth when you were off network.

[threatofrain]

> But in practice you should absolutely trust your laptop storage

As a consumer this is perhaps practical advice.

> Trusting the storage on any single machine for critical work data for any longer than necessary is unprofessional.

But if you were a customer, would you be okay with a business losing your vital data because they wanted everything on a single machine? It doesn't matter if it's a laptop, it may be a server on the cloud.

[sokoloff]

I’m a consumer and a professional. Consumer me cares more about not losing my local data a lot more than professional me does. (Professional me has a lot of other copies even if not perfect backups, I could do a good-enough restore. Consumer me can’t get back a picture of my kids when they were young if I didn’t have backups.)

[cma]

They start heavily corrupting data if left unpowered for around a year or two though, in case this gives anyone false confidence in a use case they aren't designed for.

[athms]

No they don't. This myth originated when Alvin Cox (Seagate) gave a JEDEC presentation in 2015 talking about cell corruption when the drive reaches its end of life AND is stored at abnormal temperatures. Simply leaving flash unpowered won't corrupt data, especially consumer drives. This has been debunked by Alvin Cox and all the manufacturers of flash storage.

[brokenmachine]

Do all SSDs do that? Is just powering them up every now and then enough? I mean, it's not refreshing the entire drive, is it?

[noxer]

Its nonsense. For both SSD and HDD. Although HDD do have a tendency to fail shorty after reuse if they where not used for years. This is due to lubrication decomposing afaik.

[londons_explore]

Yes, all SSD's suffer this effect.

Just plugging them in often isn't enough to make them notice and refresh all data - the SSD doesn't have a clock in so doesn't know it's been unplugged for a long time.

The real answer is "for every month an SSD is powered down, leave it powered up for a day"

[athms]

>Yes, all SSD's suffer this effect.

Wrong. Stop repeating nonsense.

[londons_explore]

Source for my claim: https://www.anandtech.com/show/9248/the-truth-about-ssd-data...

[athms]

Your source doesn't support your statement.

The original poster stated, "They start heavily corrupting data if left unpowered for around a year or two though, in case this gives anyone false confidence in a use case they aren't designed for." A person asked if all SSDs do that and you said, "Yes, all SSD's suffer this effect." This statement is wrong.

The AnandTech article even mentions the falsehoods and states it wants to clear up the confusion. A drive must meet certain conditions before data loss begins. Simply putting a backup or thumb drive in a safe for several years isn't going to cause data loss and the AnandTech article comes to this conclusion.

Here is another article from PCWorld with commentary from the authors of the 2015 JEDEC presentation that started this hysteria.

https://www.pcworld.com/article/2925173/debunked-your-ssd-wo...

[cma]

That's 2015, densities have gotten much higher since then. I'll take back heavily corrupting, I've seen it happen but I don't know that it is the norm within 1-2 years.