Bitcoin Forum

This friend I know  ::) is using winrar to encrypt his wallets with fairly long passwords.
How secure is winrars password encryption, and what's the next most convenient and more reliable form of file encryption?

WinRAR uses an ineffective encryption standard (afaik). Tell him to use truecrypt (http://truecrypt.org).

128bit AES... yeah, sure - very ineffective. NOT!

Ok goshh, they didn't use aes last time I used it.

It has changed since v3.0, so together with a strong password, it'll be secure enough for some time :)

Go with TrueCrypt.

I know this shouldn't matter, but I think it would be weird to protect something so valuable with a program everyone has on their desktop.   I am not sure why I feel like it matters to me, but it does, I can't find the logic in it yet.

Well, just in case my grandma wants to brute force it  :D

Only annoying part is that you have to create a volume that is big enough, because re-sizing isn't really possible (I've saw somewhere about someone having a 150MB+ wallet.dat file)

Just create a 1gb volume and have the entire bitcoin datadir in that.

How long is fairly long? The weak link would be a brute-force attack, and the plausibility of that will directly depend on how many passwords someone would have to try to get to his.  There already exists hardware used by law enforcement to brute force WinRAR passwords.
http://www.forensic-computers.com/TACC1441.php

7-Zip has 256 bit AES. I'm using that with an 18 char password and storing my wallet completely and permanently offline, so I'm sleeping pretty well at night ;)

Now I just have to fill it with a few BTC haha!

7-Zip uses iterated SHA-256 as its key derivation function. This is weak against hardware brute force attacks. If your password really is 18 randomish characters, you should be fine. If it's one English word with a few digits before or after it, you are theoretically vulnerable to that kind of attack.

On the bright side, you don't really have to worry about someone stealing your wallet today and then breaking it in ten years when the computing power is available to do so. Shortly before the time any encryption scheme you ever used to protect your wallet becomes vulnerable to an attack (due to increasing computing power, a newly-discovered flaw, or whatever), you can simply transfer all your BitCoins to a brand new wallet using an encryption scheme that is stronger.

WINRAR is fine... providing you use a secure password...

The password search space for a Uppercas, Lowercase, Digit, and Symbols 12 digit password is 5.46 x 10^23

That would take over 100 years at one hundred trillion guesses per second.  (10x the power of the entire bitcoin network).

useful link: https://www.grc.com/%5Chaystack.htm

Are you sure? The version I have (Ver 9.20) says AES-256. And yes, 18 random chars.

Provided you understand the difference between '!HackZl0l' (awful), '1naHTG?pw77' (just good enough for now), and '34rW0,3iviQ!' (good enough for the next 30 years for sure).

An attack would be on the weakest link which is the key derivation, not the encryption.

http://www.7-zip.org/7z.html says:
"This algorithm uses cipher key with length of 256 bits. To create that key 7-Zip uses derivation function based on SHA-256 hash algorithm. A key derivation function produces a derived key from text password defined by user. For increasing the cost of exhaustive search for passwords 7-Zip uses big number of iterations to produce cipher key from text password."

18 random characters is secure for the foreseeable future.

It is weird as a 10 digit password [a-Z][0-9][!-~] has a search space of 6.05 x 10^19 and could be cracked in 10 weeks by the Bitcoin network...  Secure passwords are much more secure than you expect.

I see! Thanks for claring that up, JoelKatz.

Make sure you pick at least one character in each group:

Lowercase: abcdefghijklmnopqrstuvwxyz
Uppercase: ABCDEFGHIJKLMNOPQRSTUVWXYZ
Number: 1234567890
Symbol: `~!@#$%^&*()-_=+\|[{]};:'",<.>/? (space)

09 char = insecure
10 char = low security
11 char = medium security
12 char = good security (good enough for your wallet)
13 char = v.good enough for anything.

or if you use a file system that supports it (ext2,3,and 4, btrfs, NTFS, UFS/BFFS, reiser, XFS, and ZFS all support sparse files, and those are basically all the file systems that matter for general purposes), create the truecrypt volume as a sparse file of some suitable large size.

That won't work, since the volume is an encrypted volume it will be completely random (as in the selected size) data. So if you create a 10GB volume it will really use 10GB even if it contains no data.

Is there anyway the files could get corrupted when encrypting? If so you will loose everything right? It has happened more than one time that I try to open a .zip or .rar-file that turn out to be corrupt. I guess you need to get a offline-copy on a USB for example as well to avoid that?

no, that is not the case.  Truecrypt supports creating sparse ("dynamic") volumes, which function exactly as i specified.  see page 37 of the truecrypt user guide.

though on further research, it appears to only be available in the windows version of truecrypt for some reason.

GPU password cracking for winrar: http://www.golubev.com/rargpu.htm (http://www.golubev.com/rargpu.htm)

19,000 passwords per second on a Radeon 5970.

That is very slow rate. Even with a small mining cluster, you will not solve 10+ char non-dictionary passwords (with upper/lower case letters, numbers and symbols) in a month.

Yup, seems you are right, you can use sparse files on NTFS partitions. Although they say that the performance will be worse (not really a problem), and that it's less secure because only the used part will be encrypted (not really a problem since an encrypted .rar will also only be the encrypted data).

The advantage of using the encryption built into the RAR format is that you can distribute an encrypted RAR archive to anyone with WinRAR, 7zip (http://net-informations.com/q/mis/7zip.html) or other common software that supports the RAR format. For your use case, this is irrelevant. Therefore I recommend using a software that is dedicated to encryption.

The de facto standard since you're using Windows was TrueCrypt. TrueCrypt provides a virtual disk which is stored as an encrypted file. Not only is this more secure than WinRAR (I trust TrueCrypt, which is written with security in mind from day 1, far more than any product whose encryption is an ancillary feature), it is also more convenient: you mount the encrypted disk by providing your password, then you can open files on the disk transparently, and when you've finished you unmount the encrypted disk. Sadly TrueCrypt is no longer in active development but it's successor VeraCrypt is. VeraCrypt is based on TrueCrypt and is compatible with the old TrueCrypt containers.

Lynda

But it can be decrypted but takes a long period of time using bruteforce!

It's also best to avoid words altogether, as "ch4r4ct3r su8st!tut!0n" alone doesn't cut it anymore.  No intelligent thief is attempting to brute force anything.  They're going to try to predict the mentality you're using when coming up with a password and use it against you.

https://www.theguardian.com/technology/2016/aug/19/password-strength-meters-security

No winrar password could be easily decrypted by special software available online. any encrypted data is easily decryptable by third party software. only 1 way encryption or encryption with a heavy key is secure upto some extent like like bitcoin wallet usage private and public key.

WinRar uses pretty standard AES, do you have any sources that claim to decrypt a rar archive with any other method than bruteforce?

"One way encryption" to secure a Bitcoin wallet? Then how are you supposed to use the wallet again?

Maybe he meant one time use encryption, like one time pads. Those are too inconvenient to use even if you plan to protect thousands of BTC.

WinRAR uses "pretty good" encryption, AES 256 bit, and the key-stretching or whatever makes it crack resistant. TrueCrypt was better. Someone made a benchmark and TC cracking speed was 700 per second, while RAR cracking speed on the same hardware and software was maybe 10,000 to 20,000 per second, using GPUs.

That's still too slow for anyone using good long passwords. Just use a randomly generated password. Anything that looks like a bitcoin address should work fine. (Yes, you'd probably have to write that down somewhere as it's pretty hard to memorize a private key.)

http://www.tomshardware.com/reviews/password-recovery-gpu,2945-6.html