Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Unpowered SSDs slowly lose data

xda-developers.com

Post reply on HN
Print view View original

301–310 of 341 posts

Re: Unpowered SSDs slowly lose data

#301

Post by userbinator »

throwaway290 wrote: post #210
userbinator wrote: post #168

Earlier quoted context omitted.

4 times less capacity but 100x or more endurance or retention at the same price looks like a great deal to me. Alternatively: do you want to have 4x more capacity at 1/100th the reliability? Plenty of people would be willing to pay for SLC mode. There is an unofficial firmware hack that enables it: https://news.ycombinator.com/item?id=40405578 1TB QLC SSDs are But why won't the manufacturers let you choose? The real…

> I have an old SLC USB drive which is only 512MB, but it's nearly 20 years old and some of the very first files I wrote to it are still intact (I last checked several months ago It's not about age of drive. It's how much time it spent without power.

Most of that time was unpowered and at room temperature (~20C).

Re: Unpowered SSDs slowly lose data

#302

Post by thenthenthen »

thenthenthen wrote: post #276
leo_e wrote: post #234

We learned this the hard way with "cold" backups stored in a literal safe. We treated NVMe drives like digital stone tablets. A year later, we tried to restore a critical snapshot and checksums failed everywhere. We now have a policy to power-cycle our cold storage drives every 6 months just to refresh the charge traps. It's terrifying how ephemeral "permanent" storage actually is. Tape is annoying to manage, but at…

What is the goldielocks area here, spinning rust? My 20 year old ide hdds seem to be… ok

*zone

Re: Unpowered SSDs slowly lose data

#303

Post by illusive4080 »

ziml77 wrote: post #128
mbreese wrote: post #107

Earlier quoted context omitted.

Honestly this is one of my favorite things about ZFS. I know that a disk scan is performed every week/month (whatever schedule). And I also know that it has verified the contents of each block. It is very reassuring in that way.

You've validated that the scrub is actually running, right? I know that the lack of a default schedule for ZFS scrubs caused Linus Media Group to lose a bunch of archived videos to bitrot.

I have zed email me every time a scrub completes. I get weekly emails because I do weekly scrubs. If I don’t see an email I know something is up!

Re: Unpowered SSDs slowly lose data

#304

Post by userbinator »

zozbot234 wrote: post #194
userbinator wrote: post #178

Earlier quoted context omitted.

No, even drives written once have started showing a rise in (correctable) errors after 2 years: https://news.ycombinator.com/item?id=43739028

That's a slight rise in ECC which is entirely expected. Flash storage can be expected to rely on error correction as part of normal functioning; it's not an abnormal condition.

Bit errors occurring after only 2 years, on flash that has been written to once, is not good at all.

Re: Unpowered SSDs slowly lose data

#305

Post by drwu »

ACCount37 wrote: post #162
userbinator wrote: post #77

One key point about retention which is not often mentioned, and indeed neither does this article, is that retention is inversely proportional to program/erase cycles and decreases exponentially with increasing temperature. Hence why retention specs are usually X amount of time after Y cycles at Z temperature. Even a QLC SSD that has only been written to once, and kept in a freezer at -40, may hold data for several de…

Because no one is willing to pay for SLC. Those QLC NAND chips? Pretty much all of them have an "SLC mode", which treats each cell as 1 bit, and increases both write speeds and reliability massively. But who wants to have 4 times less capacity for the same price?

It is not 4 times, it is 16 times. Your 1TB storages would become 64GB.

Re: Unpowered SSDs slowly lose data

#306

Post by ThePowerOfFuet »

londons_explore wrote: post #267
jeremyvisser wrote: post #218

Earlier quoted context omitted.

Here’s the thing. That SSD controller is the interface between you and those blocks. If it decides, by some arbitrary measurement, as defined by some logic within its black box firmware, that it should stop returning all blocks, then it will do so, and you have almost no recourse. This is a very common failure mode of SSDs. As a consequence of some failed blocks (likely exceeding a number of failed blocks, or perhaps…

The mechanism is usually that the SSD controller requires that some work be done before your read - for example rewriting some access tables to record 'hot' data. That work can't be done because there is no free blocks. However, no space can be freed up because every spare writable block is bad or is in some other unusable state. The drive is therefore dead - it will enumerate, but neither read nor write anything.

I don't think this is correct; it could read the flash block containing the [part of the] table in question, update it in memory, erase that block, then rewrite it into the same block.

Re: Unpowered SSDs slowly lose data

#307

Post by zigzag312 »

drwu wrote: post #305
ACCount37 wrote: post #162

Earlier quoted context omitted.

Because no one is willing to pay for SLC. Those QLC NAND chips? Pretty much all of them have an "SLC mode", which treats each cell as 1 bit, and increases both write speeds and reliability massively. But who wants to have 4 times less capacity for the same price?

It is not 4 times, it is 16 times. Your 1TB storages would become 64GB.

Wrong? You are comparing number of possible values that can be stored not the amount of bits. Drive sizes are specified in bytes not possible combinations it can store.

Re: Unpowered SSDs slowly lose data

#308

Post by dotancohen »

mbreese wrote: post #243
dotancohen wrote: post #223

Earlier quoted context omitted.

I use Debian at home, with separate boot, /, and /home/ partitions. I have no idea what type of cheap memory is stuffed into the motherboard - it's certainly not homogeneous. I do prioritise resiliency over speed, or even space. Still something I should look into? Thank you!

The servers I use ZFS on are Debian, so it’s well supported in that way. I’m pretty sure ZFS on Debian uses dkms, so if you want to try it on a data partition, it will work. Still, unless you want to tinker with something new I can’t really recommend it. Would it work? Yes. Do you need it? No. You’re probably fine with whatever FS you currently have running. ZFS works on Debian, but it’s not first-party support (due…

Thank you.

Re: Unpowered SSDs slowly lose data

#309

Post by hnarn »

leo_e wrote: post #234

We learned this the hard way with "cold" backups stored in a literal safe. We treated NVMe drives like digital stone tablets. A year later, we tried to restore a critical snapshot and checksums failed everywhere. We now have a policy to power-cycle our cold storage drives every 6 months just to refresh the charge traps. It's terrifying how ephemeral "permanent" storage actually is. Tape is annoying to manage, but at…

> We now have a policy to power-cycle our cold storage drives every 6 months

Why not tape? LTO archival tape is like 10-20 years, no? It’s also quite cheap I think

Re: Unpowered SSDs slowly lose data

#310

Post by adrian_b »

jeremyvisser wrote: post #218
lazide wrote: post #213

Earlier quoted context omitted.

Huh? The issue being discussed is random blocks, yes? If your entire drive is bricked, that is an entirely different issue.

Here’s the thing. That SSD controller is the interface between you and those blocks. If it decides, by some arbitrary measurement, as defined by some logic within its black box firmware, that it should stop returning all blocks, then it will do so, and you have almost no recourse. This is a very common failure mode of SSDs. As a consequence of some failed blocks (likely exceeding a number of failed blocks, or perhaps…

When only a number of 4 kB blocks cannot be read, if the amount of affected data is less than the amount of added redundancy the archive file can still be repaired.

For instance, if you have a 40 GB backup archive with 10% redundancy, 4 GB of data, i.e. one million 4 kB data blocks can be unreadable and you can still repair the archive and recover the complete content.

It is true that the entire SSD or HDD can become bricked. The solution for this, as I have already written in my previous comment, is to duplicate any SSD/HDD used for archival purposes, which I always do.

Post reply on HN