Combinator Attac

[jonass]

Hi,
i would like to test a rar3 hash and i have the fragments of my password.

I would like to concatenate every combination from the dictionary.

As i have seen that out of 26 lines it combinates extreme low combinations..

I expect to have even more.

From this man : https://hashcat.net/wiki/doku.php?id=combinator_attack i took for dict1 and dict2 the same file, but no effect.

[atom]

If you use small wordlists in combinator mode you better use --stdout and pipe the input back to hashcat. That makes sense for slow hashes only (but rar3 is one).

[jonass]

If you use small wordlists in combinator mode you better use --stdout and pipe the input back to hashcat. That makes sense for slow hashes only (but rar3 is one).

Hi atom,
Thank you for your reply.
Could you give me an example please how to proceed exactly now?
Thx Jonas

[atom]

./hashcat -a 1 dict1.txt dict2.txt --stdout | ./hashcat -m 13000 hash.txt

[jonass]

./hashcat -a 1 dict1.txt dict2.txt --stdout | ./hashcat -m 13000 hash.txt

Thank you for your input. Now i got the syntax.
One question.
given this words for both files:

Code:

one
house
windows
help
manual
skateboard
television

The output without piping is with this command

Code:

~/hashcat-4.0.1$ ./hashcat64.bin -a1 dict_test1.txt dict_test2.txt --stdout

is 

oneone
onehouse
onewindows
houseone
househouse
housewindows
windowsone
windowshouse
windowswindows
help one
help house
help windows
manualone
manualhouse
manualwindows
skateboardone
skateboardhouse
skateboardwindows
televisionone
televisionhouse
televisionwindows
onehelp
onemanual
oneskateboard
househelp
housemanual
houseskateboard
windowshelp
windowsmanual
windowsskateboard
help help
help manual
help skateboard
manualhelp
manualmanual
manualskateboard
skateboardhelp
skateboardmanual
skateboardskateboard
televisionhelp
televisionmanual
televisionskateboard
onetelevision
housetelevision
windowstelevision
help television
manualtelevision
skateboardtelevision
televisiontelevision

I miss words like
oneskateboardmanualtelevision

So really combination of every with every word.
Do you have a hint for me? Thank you in advance

[Skwerl23]

---

Combinator will only do 2 of each word. If you would like 4 of each. You can do the following.

Code:

hashcat -a 1 dict1.txt dict1.txt --stdout > dict1combined.txt
hashcat -a 1 dict1combined.txt dict1combined.txt --stdout | hashcat -m 13000 hash.txt

Or for the second line

Code:

hashcat -a 1 -m 13000 hash.txt dict1combined.txt dict1combined.txt

Be mindful that combining wordlist has exponential growth. So if you start doing it 4 or 5 times it will be terrabytes in size.

[jonass]

---

Combinator will only do 2 of each word. If you would like 4 of each. You can do the following.

Code:

hashcat -a 1 dict1.txt dict1.txt --stdout > dict1combined.txt
hashcat -a 1 dict1combined.txt dict1combined.txt --stdout | hashcat -m 13000 hash.txt

Or for the second line

Code:

hashcat -a 1 -m 13000 hash.txt dict1combined.txt dict1combined.txt

Be mindful that combining wordlist has exponential growth. So if you start doing it 4 or 5 times it will be terrabytes in size.

Thank you.

Finally i have 26 fragments of a password which i dont know in which order they are. Is this solveable at all?
So this is fac(26) which is quite a lot.

[Skwerl23]

---

Combinator will only do 2 of each word. If you would like 4 of each. You can do the following.

Code:

hashcat -a 1 dict1.txt dict1.txt --stdout > dict1combined.txt
hashcat -a 1 dict1combined.txt dict1combined.txt --stdout | hashcat -m 13000 hash.txt

Or for the second line

Code:

hashcat -a 1 -m 13000 hash.txt dict1combined.txt dict1combined.txt

Be mindful that combining wordlist has exponential growth. So if you start doing it 4 or 5 times it will be terrabytes in size.

Thank you.

Finally i have 26 fragments of a password which i dont know in which order they are. Is this solveable at all?
So this is fac(26) which is quite a lot.

26 fragment won't be big at all. depending on how much you're combining them.

I am not 100% sure what your question is, but the math is around 26^n x (avgwordlength x n + 1). where n is how many words in a row you think it is.
so if the average length is 6 characters, and its 8 words in a row, then you will be looking at 
26^8 x (6 + 1)
this would lead to around 10 terrabytes. HOWEVER, if you combinate to only 4, and then use combinator on the end. it's super small, and is actually about 11 megabytes. The powers of exponentials is crazy. yes running the smaller dictionary with combinate and rule is about 1/10th the speed of running the full dictionary it's self, but it is worth it. and if done right, it's more like 1/3rd the speed. so that's pretty good. and it can probably run this command in under 20 hours at a 26^8 level.  If you do some neat tricks and split the file into 2 files. and then combinate each possible pair, and run 4 instances, you can cap 4 core cpu's and do it in under 5 hours.
GPU's don't perform as well as they should this way, as they are being way underutilized when in stdin mode.

another thing to realize is that if it's 7 words, combinator can mess with you as it will always have even amounts of words.
one trick is to use it and combine a few dictionaries till you get 4 words in a dictionary.
however start your original list with a single letter word.
Then do all the combinating. once you are done you can pipe the combination stdout into hashcat, and then run a rule on attack mode 0.
make the rules
: (this rule leaves the list alone)
] (this rule deletes the last character)
deleting the last character will change delete the combinations ending in just the single letter word, allowing for odd amounts of words too.

Can you clarify your question more?