Table To File Stopped Working

[Hash-IT]

I have some old .cmd's that I used to use to make "l33t" versions of words and write them to a file.

For some reason today when I decided to make a few more I was unable to get it to work. Nothing has changed with my scripts and I have re-read the wiki about this (ironically I wrote that part of the wiki !) and I still can't work out what is wrong.

Here is the code...

Code:

hashcat-cli32.exe --attack-mode 5 --table-file a.table --table-max=15 --stdout pw.txt >> Table_Result.txt

--attack-mode 5 = Table Lookup Attack

--table-file a.table = a.table is my table file

--table-max=15 = Length of longest word to modify

--stdout pw.txt = Read from file called pw.txt

>> Table_Result.txt = Write to file called Table_Result.txt

Can someone please tell me what could have changed ?

I have replaced the old hashcat and license file with the current ones.

The only output I get in the output file is...

Quote:Try --help for more help.
Usage: hashcat [options] hashfile [mask|wordfiles|directories]

Thanks.

[atom]

should be the same issue as discussed here: https://hashcat.net/forum/thread-1016.html

[Hash-IT]

Thank you very much atom, you are of course correct, that fixed it !

I am glad you are back from your holiday, hope you had a good one.

Can I ask why we now have to write NUL NUL ? Is this a bug that will be fixed or do I have to update the wiki ?

If I am updating the wiki page I like to explain what each line of code is for, so could you please explain why this has to be added now so I can write it up ? Of course if this is a bug that will be fixed there is no need to explain.

Thank you.

[M@LIK]

It was a bug, it's fixed in hashcat-0.40b4.

[Hash-IT]

(05-30-2012 09:55 PM)M@LIK Wrote:  It was a bug, it's fixed in hashcat-0.40b4.

Thanks M@LIK I will download it now.

Edit...

Yes all working now in new beta,..... just need to change my scrips again !! Grrr !

[Hash-IT]

Thanks this does seem to be working now but it is a little different than I remember so I would like to just check if this sounds ok with others .

I have a word list with passwords no less than 8 characters long in it. I have my table file in such a way to "leet" words with simple modifications such as a=4, o=0 etc

When I run my script it all seems to be working ok but I have noticed that my GPU fan speed revs up and down in a pulse like manner. The GPU percentage on the status page can vary between 0%, 35% and up to 60%. I would have expected 100%.

My temperature is at about 60c but when I crack normally (not using pipe from hashcat and table) it s usually 71c

The c/s figure is the same as I would expect at 94k ish.

So, am I to believe the c/s figure ? It seems odd to me that I am supposedly getting the same c/s figure but using my GPU mush less.

Does this sound right ?

Thanks.

[atom]

Sounds strange. Can you give some commandline you use plus the table file etc for reproduce?

[Hash-IT]

(06-03-2012 10:13 AM)atom Wrote:  Sounds strange. Can you give some commandline you use plus the table file etc for reproduce?

Of course, thanks for showing an interest.

Before I go any further this computer is XP Pro 32bit SP3. I am using oclHashcat-plus32.exe V0.7 (cannot upgrade any further on XP). If this is a known issue with V0.7 then I understand there is no hope of help and I will have to upgrade my OS.

However you may just see something wrong with my code.

I am using ATI HD5870 with Catalyst Version 11.12. Everything else works fine for me with this set up.

My command line ...

Code:

hashcat-cli32.exe --attack-mode 5 --table-file Leet_lowercase.table --table-max=15 --stdout passlist.txt | oclHashcat-plus32.exe --hash-type 2500 -n 160 mywifi.hccap" -o Found.txt

My table file ....

Code:

a=a4@^
b=b68
c=<c(
d=d)
e=e3€
f=f
g=g69&
h=h#
i=i1!
j=j;
k=k
l=l|1
m=m
n=n
o=o0
p=p
q=q
r=r
s=s5$
t=t7+†
u=uµ
v=v
w=w
x=x%
y=y
z=z2

[atom]

I tried it and can confirm it. But it is not a bug or something. Its just that hashcat can not generate the password plains fast enough. On my 8 core bulldozer it makes ~ 280k/s. That is still faster than 180k/s my hd6990 does, but you have to take into the calculation that the results can be of every length. if you say the average password length is either 6, 7, 8 or 9 we have to divide the 280k/s by 4, makes it 70k/s. In simple words. oclHashcat-plus idles while waiting for full password plain buffer.

[Hash-IT]

Thank you for taking the time to test that atom.

Just for you I devoted all Sunday to upgrading my cracking machine to Win7 and all new drivers just for hashcat-plus ! .... If Win7 is considered upgrading !

Anyway I was going to report that this "effect" was present when using Win 7 ocl-hascat-plus v0.8 and the new beta whilst also using the latest beta of hashcat.

After seeing your explanation I understand why. To prove your point more conclusively I shall separate my word lists by line length which "should" give a different result.

I know you are already considering it and you have explained the reasons you haven't migrated the table look-up feature to ocl-hascat-plus yet but I think this confirms it does need porting to be effective.

The table look-up feature is so very useful for leet words and also toggling. I personally think it is one of your best rule options and would very much like to see it added to ocl-hascat-plus when you work out how.

Thanks again atom for testing this query out for me.