GRTGen, GRTAnalyze, GRTMerge usage

Discussion of the upcoming GPU accelerated rainbow table implementation
  • Ads

GRTGen, GRTAnalyze, GRTMerge usage

Postby pellegrino_tarantino » Wed May 23, 2012 5:34 am

Where can I find better explanations about the options of GRTGen? (e.g. number of bits, chainlenght, tableindex, numtables)?

I started with this
Code: Select all
./GRTGen-OpenCL -h MD5 -c ./charsets/charsetall -l 8 -i 0 --chainlength 50000 --numchains 1000000 -t 128 -b 128 --numtables=10

GRTCLGenerateTable: Hash: 16  Pass: 16
Table generation information:

  Hash type: MD5
  Password length: 8
  Table index: 0
  Number of chains: 1000000
  Number of tables: 10
  Chain length: 50000
  Generate seed: 2507346795
  Charset filename: ./charsets/charsetall


then I use GRTAnalyze and this is my output:
Code: Select all
./GRTAnalyze parts/MD5-len8-idx0-chr95-cl50000-sd2507346795-0-v2.part
Cannot find index file parts/MD5-len8-idx0-chr95-cl50000-sd2507346795-0-v2.part.idx
Progress: 900000 / 1000000 (90.00%)   Max bits of sameness: 39

Max bits of sameness: 39


Items in 000 bits: 6 (0.00060 %)
Items in 001 bits: 2 (0.00020 %)
Items in 002 bits: 4 (0.00040 %)
Items in 003 bits: 8 (0.00080 %)
Items in 004 bits: 16 (0.00160 %)
Items in 005 bits: 32 (0.00320 %)
Items in 006 bits: 64 (0.00640 %)
Items in 007 bits: 128 (0.01280 %)
Items in 008 bits: 256 (0.02560 %)
Items in 009 bits: 512 (0.05120 %)
Items in 010 bits: 1024 (0.10240 %)
Items in 011 bits: 2048 (0.20480 %)
Items in 012 bits: 4096 (0.40960 %)
Items in 013 bits: 8192 (0.81920 %)
Items in 014 bits: 16384 (1.63840 %)
Items in 015 bits: 32768 (3.27680 %)
Items in 016 bits: 65461 (6.54610 %)
Items in 017 bits: 125262 (12.52620 %)
Items in 018 bits: 190010 (19.00100 %)
Items in 019 bits: 198447 (19.84470 %)
Items in 020 bits: 150596 (15.05960 %)
Items in 021 bits: 94350 (9.43500 %)
Items in 022 bits: 52812 (5.28120 %)
Items in 023 bits: 28318 (2.83180 %)
Items in 024 bits: 14502 (1.45020 %)
Items in 025 bits: 7247 (0.72470 %)
Items in 026 bits: 3703 (0.37030 %)
Items in 027 bits: 1854 (0.18540 %)
Items in 028 bits: 990 (0.09900 %)
Items in 029 bits: 467 (0.04670 %)
Items in 030 bits: 223 (0.02230 %)
Items in 031 bits: 119 (0.01190 %)
Items in 032 bits: 56 (0.00560 %)
Items in 033 bits: 20 (0.00200 %)
Items in 034 bits: 10 (0.00100 %)
Items in 035 bits: 5 (0.00050 %)
Items in 036 bits: 1 (0.00010 %)
Items in 037 bits: 1 (0.00010 %)
Items in 038 bits: 4 (0.00040 %)
Items in 039 bits: 1 (0.00010 %)
Items in 040 bits: 0 (0.00000 %)
Items in 041 bits: 0 (0.00000 %)
Items in 042 bits: 0 (0.00000 %)
Items in 043 bits: 0 (0.00000 %)
Items in 044 bits: 0 (0.00000 %)
Items in 045 bits: 0 (0.00000 %)
Items in 046 bits: 0 (0.00000 %)
Items in 047 bits: 0 (0.00000 %)
Items in 048 bits: 0 (0.00000 %)
Items in 049 bits: 0 (0.00000 %)
Items in 050 bits: 0 (0.00000 %)
Items in 051 bits: 0 (0.00000 %)
Items in 052 bits: 0 (0.00000 %)
Items in 053 bits: 0 (0.00000 %)
Items in 054 bits: 0 (0.00000 %)
Items in 055 bits: 0 (0.00000 %)
Items in 056 bits: 0 (0.00000 %)
Items in 057 bits: 0 (0.00000 %)
Items in 058 bits: 0 (0.00000 %)
Items in 059 bits: 0 (0.00000 %)
Items in 060 bits: 0 (0.00000 %)
Items in 061 bits: 0 (0.00000 %)
Items in 062 bits: 0 (0.00000 %)
Items in 063 bits: 0 (0.00000 %)
Items in 064 bits: 0 (0.00000 %)
Items in 065 bits: 0 (0.00000 %)
Items in 066 bits: 0 (0.00000 %)
Items in 067 bits: 0 (0.00000 %)
Items in 068 bits: 0 (0.00000 %)
Items in 069 bits: 0 (0.00000 %)
Items in 070 bits: 0 (0.00000 %)
Items in 071 bits: 0 (0.00000 %)
Items in 072 bits: 0 (0.00000 %)
Items in 073 bits: 0 (0.00000 %)
Items in 074 bits: 0 (0.00000 %)
Items in 075 bits: 0 (0.00000 %)
Items in 076 bits: 0 (0.00000 %)
Items in 077 bits: 0 (0.00000 %)
Items in 078 bits: 0 (0.00000 %)
Items in 079 bits: 0 (0.00000 %)
Items in 080 bits: 0 (0.00000 %)
Items in 081 bits: 0 (0.00000 %)
Items in 082 bits: 0 (0.00000 %)
Items in 083 bits: 0 (0.00000 %)
Items in 084 bits: 0 (0.00000 %)
Items in 085 bits: 0 (0.00000 %)
Items in 086 bits: 0 (0.00000 %)
Items in 087 bits: 0 (0.00000 %)
Items in 088 bits: 0 (0.00000 %)
Items in 089 bits: 0 (0.00000 %)
Items in 090 bits: 0 (0.00000 %)
Items in 091 bits: 0 (0.00000 %)
Items in 092 bits: 0 (0.00000 %)
Items in 093 bits: 0 (0.00000 %)
Items in 094 bits: 0 (0.00000 %)
Items in 095 bits: 0 (0.00000 %)
Items in 096 bits: 0 (0.00000 %)
Items in 097 bits: 0 (0.00000 %)
Items in 098 bits: 0 (0.00000 %)
Items in 099 bits: 0 (0.00000 %)
Items in 100 bits: 0 (0.00000 %)
Items in 101 bits: 0 (0.00000 %)
Items in 102 bits: 0 (0.00000 %)
Items in 103 bits: 0 (0.00000 %)
Items in 104 bits: 0 (0.00000 %)
Items in 105 bits: 0 (0.00000 %)
Items in 106 bits: 0 (0.00000 %)
Items in 107 bits: 0 (0.00000 %)
Items in 108 bits: 0 (0.00000 %)
Items in 109 bits: 0 (0.00000 %)
Items in 110 bits: 0 (0.00000 %)
Items in 111 bits: 0 (0.00000 %)
Items in 112 bits: 0 (0.00000 %)
Items in 113 bits: 0 (0.00000 %)
Items in 114 bits: 0 (0.00000 %)
Items in 115 bits: 0 (0.00000 %)
Items in 116 bits: 0 (0.00000 %)
Items in 117 bits: 0 (0.00000 %)
Items in 118 bits: 0 (0.00000 %)
Items in 119 bits: 0 (0.00000 %)
Items in 120 bits: 0 (0.00000 %)
Items in 121 bits: 0 (0.00000 %)
Items in 122 bits: 0 (0.00000 %)
Items in 123 bits: 0 (0.00000 %)
Items in 124 bits: 0 (0.00000 %)
Items in 125 bits: 0 (0.00000 %)
Items in 126 bits: 0 (0.00000 %)
Items in 127 bits: 0 (0.00000 %)


But whe i start GRTMerge this way
Code: Select all
./GRTMerge --buildfull parts/MD5-len8-idx0-chr95-cl50000-sd*.part --bits=48 -o /media/Win7/GRT/

i get this:
Code: Select all
Table version:   2
Hash:            MD5
Password length: 8
Table index:     0
Chain length:    50000
Num chains:      1000000
Perfect table:   No
Charset length:  95
Charset:          !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQR
                 STUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~



Checking Files... 10 / 11   

Total to merge: 11
Output dir: /media/Win7/GRT/
ERROR: Must set bitsInHash + bitsInPassword to a multiple of 8 (byte aligned)!
bitsInhash: 48
bitsInPassword: 53
Try 43 or 51 for bits in hash.


what results do i need to correctly use GRTMerge?

Thank you
pellegrino_tarantino
 
Posts: 16
Joined: Fri May 11, 2012 11:33 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby Bitweasil » Wed May 23, 2012 2:22 pm

http://cryptohaze.com/wiki/index.php/Cr ... bow_Tables
http://cryptohaze.com/wiki/index.php/Rainbow_Tables

I thought I had a writeup on generating length 6 tables, but I can't find it...
Bitweasil
Site Admin
 
Posts: 912
Joined: Tue Jan 20, 2009 4:26 pm

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby pellegrino_tarantino » Wed May 23, 2012 7:18 pm

Bitweasil wrote:http://cryptohaze.com/wiki/index.php/Cryptohaze_GPU_Rainbow_Tables
http://cryptohaze.com/wiki/index.php/Rainbow_Tables


Ok, i read both explanations, so what difference if i create 100 or 200 tables or more? and what about the --numchains? these values are better if higher? I tuned chainlenght to 200000.
here's what i did:
Code: Select all
./GRTGen-OpenCL -h MD5 -c ./charsets/charsetall -l 8 -i 0 --chainlength 200000 --numchains 200000 --numtables 100 -t 128 -b 128 --bfi_int


I thought I had a writeup on generating length 6 tables, but I can't find it...


Yes, the example you're talking about is here http://cryptohaze.com/wiki/index.php/GRTGen#Example

Example
GRTGen-CUDA -h MD5 -c ./charsets/charsetall -l 6 -i 0 --chainlength 50000 --numchains 10000000 -t 128 -b 128 -m 500
pellegrino_tarantino
 
Posts: 16
Joined: Fri May 11, 2012 11:33 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby pellegrino_tarantino » Wed May 23, 2012 9:26 pm

is there any way to force GRTCrack to exclude CPU (e.g. --nocpu)? it uses CPU during "search" step, and GPU only during the "regen" step
pellegrino_tarantino
 
Posts: 16
Joined: Fri May 11, 2012 11:33 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby Bitweasil » Fri May 25, 2012 3:47 am

pellegrino_tarantino wrote:is there any way to force GRTCrack to exclude CPU (e.g. --nocpu)? it uses CPU during "search" step, and GPU only during the "regen" step


No - the CPU side is how you search tables. It's using the hard drive.

If you don't have a few GPU years of time, you're not going to be able to generate your own set of len8 tables.
Bitweasil
Site Admin
 
Posts: 912
Joined: Tue Jan 20, 2009 4:26 pm

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby pellegrino_tarantino » Wed May 30, 2012 7:50 pm

Bitweasil wrote:If you don't have a few GPU years of time, you're not going to be able to generate your own set of len8 tables.


ok, that was the answer i wanted to hear, so i can stop wasting my time... ;)
pellegrino_tarantino
 
Posts: 16
Joined: Fri May 11, 2012 11:33 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby mailmuncher2000 » Mon Sep 23, 2013 1:36 am

I had a similar questions which I haven't found an answer for in the forums or the wiki or elsewhere. Can you tell me if my understanding of the parameter "bits" is correct with reference to the following tools:

GRTAnalyze: When I run GRTAnalyze on a part which I just generated it tells me the number of bits of sameness. Q1. Does this refers to the number of bits of sameness of the hash or of the hash+password?

GRTMerge: So according to the wiki the "bits" parameter takes bits of hash to be kept. Also different "parts" have different bits of sameness. For eg: "xxx1-0-v2.part" has 48 bits of sameness where as "xxx2-0-v2.part" has 49 bits of sameness. Q2. Is my assumption correct that I need to use the bits of sameness I got from GRTanalyze above and more specifically the highest (say 49) of the many "parts" that are to be merged? I get this output sometimes:

Code: Select all
ERROR: Must set bitsInHash + bitsInPassword to a multiple of 8 (byte aligned)!
bitsInhash: 49
bitsInPassword: 53
Try 43 or 51 for bits in hash.


Q3. So I need to ensure that the bits parameter value + 53 = a multiple of 8 correct? Hence the suggestion to use 43 or 51. Q4. Another question is what will happen if I choose to use 43. Does this mean I will loose some data from the part/table i.e. I will only be storing the chains which can be captured using the bits I have requested? Q5. So looking at the output of GRTAnalyze could I make a decision on whether to do this based on the graph....i.e. items in 43 bits is a very small percentage then I am not losing much....correct?

GRTIndex: So lastly when I use GRTIndex the bits I specify here are for searching only... I am allowed to enter 0 to 32 bits for indexing. Q6. So what are good number of bits to enter here and what aspects of the process am I speeding up here by adding more bits? Q7. Is my assumption that adding more bits to the index increases the size of the index file, but also narrows down the location of the target chain?

Hopefully I have made this easy for someone to answer with a yes or a no(plus small explanation) for each question. Thanks.
mailmuncher2000
 
Posts: 2
Joined: Mon Sep 23, 2013 12:48 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby mailmuncher2000 » Sat Sep 28, 2013 4:21 pm

Anyone ? :cry:
mailmuncher2000
 
Posts: 2
Joined: Mon Sep 23, 2013 12:48 am

Re: GRTGen, GRTAnalyze, GRTMerge usage

Postby wajahat30 » Sat Sep 20, 2014 11:40 am

made a nice post... :D
watson
wajahat30
 
Posts: 1
Joined: Sat Sep 20, 2014 11:39 am


Return to GPU Rainbow Tables

Who is online

Users browsing this forum: No registered users and 2 guests

cron