Email or username:

Password:

Forgot your password?
All posts nixCraft's posts Post Back to profile
Top-level
4censord

@nazokiyoubinbou @nixCraft there is a point of overwrites where every flash cell has been changed. But, that costs loads of disk wear. Instead, one can issue a data secure erase command to drop all data, that even restores some write performance

12 March at 21:54 | Wall-to-wall | Open on unfug.social
9 comments
Nazo

@4censord @nixCraft I know about erasing blocks, yeah. But I presume we're talking about extreme measures when someone goes as far as to actually drill holes in it. In theory it's possible to recover old data from erased cells -- or at least so I've heard. Presumably the idea here is to ensure that is no longer possible.

Sure is wasteful as heck to drill it IMO.

12 March at 21:56 | Open on mastodon.social
4censord
Nazo

@4censord @nixCraft This is what I was talking about, yes.

12 March at 22:00 | Open on mastodon.social
Zimmie

@4censord @nazokiyoubinbou @nixCraft There are two forms of secure erase for SSDs.

One involves flushing the key used for the wear leveling encryption. This is extremely fast, but it doesn’t actually erase the data; only makes it unreadable. Due to this, it doesn’t directly help with performance, so it is commonly followed by the second, older type.

The older type involves a voltage spike for about a tenth of a second. This completely wipes all data stored in the chip’s pages. This is the one which restores write performance, since all the pages are blank again (no mode read/erase/program cycle). Since this is a chip-level thing, it even wipes the data stored in spared pages.

@4censord @nazokiyoubinbou @nixCraft There are two forms of secure erase for SSDs.

One involves flushing the key used for the wear leveling encryption. This is extremely fast, but it doesn’t actually erase the data; only makes it unreadable. Due to this, it doesn’t directly help with performance, so it is commonly followed by the second, older type.

12 March at 23:11 | Open on infosec.exchange
Nazo

@bob_zim @4censord @nixCraft Oh, I misunderstood. I thought that was referring to a simple block erase like what TRIM uses. That restores performance in the same basic way, but this mechanism takes it further by more thoroughly clearing the blocks I see.

12 March at 23:14 | Open on mastodon.social
Zimmie

@nazokiyoubinbou @4censord @nixCraft Technically, TRIM erases pages. Pages on a flash chip can contain several blocks. Erasing a single flash page is basically the voltage spike, but to only the one page and for a shorter time. The chip-level erase lasts longer in part to make sure everything gets fully saturated.

Different trigger and different scale of effect, but it’s the same mechanism.

12 March at 23:24 | Open on infosec.exchange
Nazo

@bob_zim @4censord @nixCraft Yeah, I get you.

This makes a lot more sense than drilling. Then maybe they can be resold or repurposed instead of just being wasteful.

12 March at 23:25 | Open on mastodon.social
Robbie 🇧🇪 :tux:

@4censord
Also especially with mlc flash chips it is impossible to know which bits were stored once the cell state has been altered or reset. There is no physical magnetic memory effect like on harddrives
@nazokiyoubinbou @nixCraft

14 March at 7:58 | Open on mastodon.green
Frank Heijkamp

@4censord @nazokiyoubinbou @nixCraft This only works on SSD drives that supports this feature.

13 March at 6:49 | Open on mastodontech.de
Go Up