Does installing GrapheneOS wipe all data permanently

Johnnyloans

Experts advise that writing the entire disk doesn't fill the entire disk and to not rely solely on this method

Upstate1618

Johnnyloans No. A factory reset deletes nothing save for the encryption key. I explained this all above.

Let's say that I create a secret language by moving the last letter in every word to the beginning of the word. I wrote how to decypher my writings. I am 5 years old and a bully steals my diary. I destroy the instructions. 15 year old Timmy will figure out how to read my secret encryption. The data exists.
This is SHA1's history and SHA256's future.

A factory reset indeed deletes everything irrevocably. After factory reset the salt used to derive encryption key is gone. It's impossible to decrypt the data afterwards.

Johnnyloans

Please raccoondad, anger and hubris are the antithesis of growth and learning. There's more to learn than c# and data structures.

raccoondad SHA1 isn't an encryption algorithm, its a hashing algorithm.

quillet
Copy my text into word and replace sha1 with rc4 if it'll help retention. I have never used SHA1.
"its" -> "it's"

raccoondad holy hell learn basic cryptographic theory and NIST standards and come back to me.

argumentum ad verecundiam
Belligerent

raccoondad I dont have a reason to argue with the uneducated.

baseless ad hominem

The situation is rather plain when i back up my claims for a 2nd time and your only response is: this 1 word is incorrect, please adjust.
I'd expect a defense panel to offer up more than this. You wrote over 200 words and even googled DoD #### for this but can't offer up anything in direct response.
It's dangerous to pretend SHA256 will be impossible to decypher in the future.

Johnnyloans 100%, this will actually lower your devices life, as the SSD can only be written into so many times.

Johnnyloans 10h/1.5E6h = 0.0007% usage
Johnnyloans my Pixel 6 is 4.5 years old and GOS says it has 92% SSD life remaining. We can extrapolate this to an estimated 56 year lifespan

There was no point in further replying to upstate but just to show how silly blidnly agreeing is and it's humorous:

Upstate1618 A factory reset indeed deletes everything irrevocably.

encrypt a drive with RC4, delete the key then get back to me.

Upstate1618 After factory reset the salt used to derive encryption key is gone. It's impossible to decrypt the data afterwards.

Salt is upstream to the key, not downstream. Salt isn't a secret weapon that the key pulls out last second in a movie.
This quote is similar to saying: "I painted my product blue, making it structurally sound."
Salt doesn't strengthen the algorithm. Salt is primarily a rainbow table defense.

Upstate1618 It's security theatre and gives people false sense of security. Such suggestion distracts people from real security best practice.

"A false sense of security" implies that security is lower.
Phone > flash grapheneOS = safe
Phone > Add a 20GB mp4 file > flash grapheneOS = false sense of security

So grapheneOS security is full-proof except for when i record a 30 minute video of my puppy, got it.

Upstate1618

Johnnyloans recommended against

It's security theatre and gives people false sense of security. Such suggestion distracts people from real security best practice.

raccoondad

Johnnyloans This is SHA1's history and SHA256's future.

No? SHA1 and MD5 hashes can't be reverted back to their input data. That's not how hashing works. Its not cryptographic reality

SHA1 isn't an encryption algorithm, its a hashing algorithm.

Johnnyloans

raccoondad

Youre right.

You should try that ^ some time.
It's okay to be wrong or even silly sometimes raccoon dad.

My points still stand.

Take care

Johnnyloans

raccoondad There is no 'replacing' data [on a storage device.]

This is a failure to grasp the rudimentary concepts of "maximum capacity of a drive" as well as what "a full drive" means.

raccoondad No? SHA1 and MD5 hashes can't be reverted back to their input data.

Encrypted files can decrypted once an algorithm is broken enough. This is a failure to understand basic cryptography history.

raccoondad You can't reverse SHA hashes, 1 or 256, without brute force.

This is a basic misunderstanding of the concepts: "broken algorithm" and "security of today isn't perfect forever." It'll either be so broken that brute forcing is either not needed or perfectly realistic to do because the possibilities get cut down.

I implemented: "don't attribute to malice what can be explained by ignorance"

raccoondad

Johnnyloans My points still stand.

No, it doesn't. Nothing you have said abides in reality. You can't reverse SHA hashes, 1 or 256, without brute force.

Johnnyloans

raccoondad

Oh? then please educate us. Does deleting the key delete the 1s and 0s off of my drive after all?

Offer more than semantics

raccoondad

Johnnyloans holy hell learn basic cryptographic theory and NIST standards and come back to me. I'm done with this, I dont have a reason to argue with the uneducated.

kaisernatron

Johnnyloans https://grapheneos.org/faq#encryption

The OS stores a high entropy random value as the Weaver token on the secure element (Titan M on Pixels) and uses it as another input for key derivation. The Weaver token is stored alongside a Weaver key derived by the OS from the password token. In order to retrieve the Weaver token, the secure element requires the correct Weaver key. Weaver also provides reliable wiping of data since the secure element can reliably wipe a Weaver slot. Deleting a profile will wipe the corresponding Weaver slot and a factory reset of the device wipes all of the Weaver slots. The secure element also provides insider attack resistance preventing firmware updates before authenticating with the owner profile.

I'm not tech savvy but this is what I inferred from the FAQ. Both the key and token are needed to access the data. The key is wiped along with the token, so even if you manage to get the key one day, the phone won't be able to access the data. Since the Pixel with stock OS still has Titan M, factory resetting it should also do the same as it does on GOS. Please correct me if I'm wrong

Upstate1618

Johnnyloans encrypt a drive with RC4

Not true. Android does not use RC4. I doubt you don't even know what method Android uses for FBE. The whole thing has nothing to do with RC4 or SHA1 actually.

Johnnyloans implies that security is lower.

Not true. Security theater means the suggestion is meaningless. It's a waste of time and effort.

Johnnyloans Salt doesn't strengthen the algorithm. Salt is primarily a rainbow table defense.

You are misrepresenting what I'm saying. I said without salt the data can't be decrypted.

Johnnyloans They wait 5, 10, 20 years until the encryption of today is eventually feasible to reverse. They then read the contents.

The encryption method Android uses is very strong that can't be exploited in hundreds of years. Even if the method is explored in 2027, your suggestion of putting large files into the phone is still a bad one cause GrapheneOS data would also be decrypted. The right thing to do is avoiding using any ohone and PC in that case.

Johnnyloans

kaisernatron

factory resetting, duress pin, deleting the key are all perfect solutions for 99.99999999% of people.

A resourceful organization could remove chips that actually store the data from your phone and put them onto another circuit board which allows them to access the storage. They then clone every 1 and 0 on the drive. They keep the original if they'd like. They wait 5, 10, 20 years until the encryption of today is eventually feasible to reverse. They then read the contents.

Upstate1618

My point is your suggestion is awfully bad whether the encryption method is exploited or not. No one should do that under any circumstances unless they just want to kill their time.

DeletedUser708

raccoondad sorry for the late reply- 1 SSD cycle isnt that deep though right? Dont they have like super high TBW guaranteed?

EverettHitch

DeletedUser708 talk English please.

@all
I have been trying to find out more about GrapheneOS's "zero on free". Do you have any pointers?

Johnnyloans

DeletedUser708 1 SSD cycle isnt that deep though right? Dont they have like super high TBW guaranteed?

You're right. Raccoon doesn't know what he's talking about and has too much of an ego to admit to his hasty reply.
https://en.wikipedia.org/wiki/Harvest_now,_decrypt_later

His only rebuttal to me was I got 1 word wrong. He's a fantastic associate editor.

Johnnyloans Here is a 5-year-old, middle-of-the-road SSD's datasheet:
https://download.semiconductor.samsung.com/resources/data-sheet/Samsung_NVMe_SSD_980_Data_Sheet_Rev.1.1.pdf

Warranty
MTBF = 1,500,000 hours
TBW = 600 x capacity

Let's be generous to your arguement and say businesses clamp warranty claims to: <10%. 90% of drives will exceed these writes and hours.

Let's give a generous 10 hours write time.

10h/1.5E6h = 0.0007% usage
1.3c/600c = 0.002% usage

Real-world usage is less than this as explained above.

Speaking of, my Pixel 6 is 4.5 years old and GOS says it has 92% SSD life remaining. We can extrapolate this to an estimated 56 year lifespan. If you'll excuse my boldness, I'd say I'm above avg in terms of writes.

Johnnyloans

EverettHitch I have been trying to find out more about GrapheneOS's "zero on free". Do you have any pointers?

This is: if any application doesn't need a bit of memory anymore, GOS places zeros into the entire section of memory.

Great for removing secrets. Cellebrite read iOS unlocking secrets from memory.

Only thing I've seen is this doc:
https://grapheneos.org/features#exploit-mitigations
Has the phrase "zero-on-free" a bit further down.

Nothing to do with permanent storage. It's impossible to zero out small sections of an SSD unless a manufacturer has implemented this into the controller which I haven't heard of.

Novalissoide

Johnnyloans but read about the Weaver token:
https://grapheneos.org/faq#encryption

Think of deletion in modern encryption is like destroying not only the keys, but also their locks. So no recovery, even when harvesting and going quantum.

Johnnyloans

Novalissoide but read about the Weaver token:
https://grapheneos.org/faq#encryption

A high level way of explaining this:

Scenario 1:

I have a password to my phone's data but I first must go talk to Titan M

I give the secret passcode and then a secret handshake. then he tells me a secret hardware code that only he can read.
Then I finally get the full key!: password123

All of these steps don't make my password123 SUPER strong or anything.

Make the above even simpler with Scenario 2:

"My phone can never be unlocked because the cops don't know that the phone MUST be held upside down while you type in the pin"

These are hurdles to obtaining the final key. It slows attackers. It doesnt add entropy to the key or pin or make the encryption stronger.

we wait until the encryption used is weak

If the encryption is weak/broken (let's say it's an iphone 1), we can skip all of that bullshit and just say.

F%&@ it. I'm done with the handshakes and flipping the phone in every direction.

Transfer the phone's storage chips to this PC SSD. We plug it into a computer then we chain together various known weaknesses in the encryption algorithm which lowers the entropy of the key drastically. We can brute force it in no time.

A higher level of what I just said is this:
Grandma was rather forgetful. Her password must be 1 of a few things. Only a few guesses now and we will be in.

All of this hashing and what not with the weaver key doesnt add entropy. The final password is the final password.
Password123

These algorithms won't get stronger as time passes.

File data is encrypted with AES-256-XTS and file names with AES-256-CTS.

Johnnyloans

Novalissoide Think of deletion in modern encryption is like destroying not the keys, but the locks. So no recovery, even when harvesting and going quantum.

The keys and locks are gone. The storage box remains and entropy reduction exists.

Novalissoide

Johnnyloans give the secret passcode and then a secret handshake. then he tells me a secret hardware code that only he can read.
Then I finally get the full key!: password123

But I think this part of the analogy is what I would object to. In that the key that the 'gatekeeper' holds is vastly longer than 123password or anything like a 'human' password.

« Previous Page Next Page »