Add new algorithm: Woltlab Burning Board with short salts #957

Open
Francisko82 opened this Issue Jan 8, 2017 · 4 comments

Comments

Projects
None yet
4 participants
@Francisko82

Algorithm is sha1($salt.sha1($salt.sha1($pass))), difference from WBB3 is in that salt length is fixed in 9 chars.

It is actually a variant of Woltlab Burning Board and it has been seen on some new sites (I have 11k private hashes of this type).

Salt length: 9 chars
Password length: 4-20 chars or same as WBB3

Sample hash/salt/plain pairs:
4b98d552f3a2fbbff35de2a9213b680c107bbeda:b7aa88903:chaosb b8395af3e5c3a26b5670d0ca665a778b2bcc4645:80bb20214:doctork1 798fcb9a474b653898f6b29912b2d4457256bb17:4ad8dcbda:dioxide1

atom got me to open this thread from here: https://hashcat.net/forum/thread-6166.html
This makes me hope this new mode can be added in future! Thanks!

@Francisko82 Francisko82 changed the title from Add new algorithm: Woltlab Burning Board with short hashes to Add new algorithm: Woltlab Burning Board with short salts Jan 8, 2017

@hops

This comment has been minimized.

Show comment
Hide comment
@hops

hops Jan 25, 2017

Contributor

That looks like hashes from OpenCart and is supported since hashcat v3.10
https://github.com/hashcat/hashcat/blob/master/docs/changes.txt#L334

Contributor

hops commented Jan 25, 2017

That looks like hashes from OpenCart and is supported since hashcat v3.10
https://github.com/hashcat/hashcat/blob/master/docs/changes.txt#L334

@Francisko82

This comment has been minimized.

Show comment
Hide comment
@Francisko82

Francisko82 Jan 28, 2017

Hops, that's absolutely true and thanks for your awesome reply! I have searched a lot before posting but didn't find this info... even atom has told me to request a new mode here.

I still cannot use my Iris GPU in OSX on this hash, but that's another story...
Btw if anyone could provide me with the correct line to edit the hashcat.hctune file to work with Iris GPU on OSX with this hash mode, I'd be very grateful!

Francisko82 commented Jan 28, 2017

Hops, that's absolutely true and thanks for your awesome reply! I have searched a lot before posting but didn't find this info... even atom has told me to request a new mode here.

I still cannot use my Iris GPU in OSX on this hash, but that's another story...
Btw if anyone could provide me with the correct line to edit the hashcat.hctune file to work with Iris GPU on OSX with this hash mode, I'd be very grateful!

@hops

This comment has been minimized.

Show comment
Hide comment
@hops

hops Feb 1, 2017

Contributor

I was the one who added this to hashcat that's why I've noticed the scheme and especially the password length restriction.
Btw. OpenCart uses PHP's mb_strlen() for the length check. This means that a multibyte UTF-8 encoded character counts only as one.
That's the reason I didn't enforce any length checks in hashcat for this hashmode.

Since this has been resolved, please close this issue.

Contributor

hops commented Feb 1, 2017

I was the one who added this to hashcat that's why I've noticed the scheme and especially the password length restriction.
Btw. OpenCart uses PHP's mb_strlen() for the length check. This means that a multibyte UTF-8 encoded character counts only as one.
That's the reason I didn't enforce any length checks in hashcat for this hashmode.

Since this has been resolved, please close this issue.

@philsmd

This comment has been minimized.

Show comment
Hide comment
@philsmd

philsmd Feb 2, 2017

Member

Well, it would be great if somebody can investigate what is going on here about the salt length...
as far as we know Woltlab Burning Board 3 uses exactly 40 bytes salt.
Newer versions of WBB (4.x) use a totally different algorithm: #73

-m 8400 = WBB3 (Woltlab Burning Board) already supports the algorithm sha1($salt.sha1($salt.sha1($pass))), but of course (mostly because of optimizations) restricts the salt to exactly 40 bytes.
So why now the salt can be shorter? Is this restriction wrong? ... or is there even a third version of the WBB algorithm?
We can't just silently close this issue, something needs to be done (either we need to do some changes to -m 8400 or we need to add some additional hash type and information to the help such that other users can crack those hashes too).
It would be great if someone can do some reaseach/investigation about this missing information and resolve these doubts.
Thanks

Member

philsmd commented Feb 2, 2017

Well, it would be great if somebody can investigate what is going on here about the salt length...
as far as we know Woltlab Burning Board 3 uses exactly 40 bytes salt.
Newer versions of WBB (4.x) use a totally different algorithm: #73

-m 8400 = WBB3 (Woltlab Burning Board) already supports the algorithm sha1($salt.sha1($salt.sha1($pass))), but of course (mostly because of optimizations) restricts the salt to exactly 40 bytes.
So why now the salt can be shorter? Is this restriction wrong? ... or is there even a third version of the WBB algorithm?
We can't just silently close this issue, something needs to be done (either we need to do some changes to -m 8400 or we need to add some additional hash type and information to the help such that other users can crack those hashes too).
It would be great if someone can do some reaseach/investigation about this missing information and resolve these doubts.
Thanks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment