Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
oclhc+ bug?
02-23-2011, 08:52 PM
Post: #1
oclhc+ bug?
Somehow I have hit a wall with oclhc+ which looks like a bug. As I did not find any pw so far, I set up a test scenario and I did the same hashfile and wordlist with both hc 0.36 and oclhc+ 0.03. While hc (correctly) finds 13 passes, oclhc+ only finds 2.

For testing purposes, I am not using a ruleset with either of the programs.

Please see full cmdline and output below.

Am I doing something wrong?

Thanks
J+

#############HC OUTPUT:

# hashcat-0.36/hashcat-cli64.bin -m 1100 -o ../data/fnd ../data/HASHES.ALL.hc8 ../data/passdic
Initializing with 8 threads and 32mb segment-size...

NOTE: press enter for status-screen

Added hashes from file ../data/HASHES.ALL.hc8: 46 (44 salts)
Wordlist..: ../data/passdic
Index.....: 1/1 (segment), 19 (words), 140 (bytes)
Recovered.: 13/46 hashes, 13/44 salts
Speed/sec.: - plains, - words
Progress..: 19/19 (100.00%)

#############OCLHC+ OUTPUT:

# oclHashcat+-0.03/oclHashcat+64.bin -m 1100 -o ../data/fnd ../data/HASHES.ALL.hc8 ../data/passdic
oclHashcat+ v0.03 starting...

Hashes: 46
Bitmaps: 9 bits, 512 entries, 0x000001ff mask, 2048 bytes
Rules: 1
Platform: ATI compatible platform found
Device #1: Cayman, 1024MB, 0Mhz, 22MCU
Device #2: Cayman, 1024MB, 0Mhz, 22MCU
NOTE: gpu-accel auto-adjusted to: 8
Device #1: Kernel oclHashcat+-0.03/kernels/4098/m1100.Cayman.64.kernel (1136096 bytes)
Device #2: Kernel oclHashcat+-0.03/kernels/4098/m1100.Cayman.64.kernel (1136096 bytes)

Starting attack in wordlist_mode...

p: 0/46, cs: 16, cr: 0, cl: 4, rt: 7.33ms, s: 4.37k/s
p: 0/46, cs: 32, cr: 0, cl: 4, rt: 4.16ms, s: 7.69k/s
p: 1/46, cs: 2, cr: 0, cl: 5, rt: 5.47ms, s: 5.48k/s
p: 1/46, cs: 18, cr: 0, cl: 5, rt: 3.12ms, s: 5.14k/s
p: 1/46, cs: 34, cr: 0, cl: 5, rt: 3.15ms, s: 5.08k/s
p: 1/46, cs: 5, cr: 0, cl: 6, rt: 4.29ms, s: 16.56k/s
p: 1/46, cs: 21, cr: 0, cl: 6, rt: 3.23ms, s: 59.39k/s
p: 2/46, cs: 37, cr: 0, cl: 6, rt: 3.47ms, s: 55.32k/s
p: 2/46, cs: 8, cr: 0, cl: 7, rt: 3.85ms, s: 27.03k/s
p: 2/46, cs: 25, cr: 0, cl: 7, rt: 2.66ms, s: 6.03k/s
p: 2/46, cs: 41, cr: 0, cl: 7, rt: 2.73ms, s: 5.87k/s
p: 2/46, cs: 12, cr: 0, cl: 8, rt: 3.92ms, s: 7.14k/s
p: 2/46, cs: 29, cr: 0, cl: 8, rt: 2.74ms, s: 11.66k/s
p: 2/46, cs: 46, cr: 0, cl: 8, rt: 2.72ms, s: 11.74k/s
p: 2/46, cs: 16, cr: 0, cl: 13, rt: 3.90ms, s: 4.11k/s
p: 2/46, cs: 33, cr: 0, cl: 13, rt: 2.73ms, s: 5.85k/s

Processed 19 words in 111.91ms = 0.17k/s
Find all posts by this user
02-23-2011, 09:34 PM
Post: #2
RE: oclhc+ bug?
can you try with md5 instead of dcc? dcc had problems in oclhc, too
Visit this user's website Find all posts by this user
02-24-2011, 05:08 PM
Post: #3
RE: oclhc+ bug?
(02-23-2011 09:34 PM)atom Wrote:  can you try with md5 instead of dcc? dcc had problems in oclhc, too

thanks, sure, will do so at night when I have repaired ssh access to the box which I wrecked yesterday Smile
Find all posts by this user
02-25-2011, 10:15 AM
Post: #4
RE: oclhc+ bug?
(02-24-2011 05:08 PM)acubens Wrote:  
(02-23-2011 09:34 PM)atom Wrote:  can you try with md5 instead of dcc? dcc had problems in oclhc, too

thanks, sure, will do so at night when I have repaired ssh access to the box which I wrecked yesterday Smile


OK tried it with -m 0 and the example files, and this also seems strange. hashcat states it recovers all (102) hashes, oclhc+ only a third (33). If this is relevan, I am using two HD6950 flashed to HD6970.

#################hc#############
hashcat-0.36/hashcat-cli64.bin -m 0 hashcat-0.36/examples/A0.M0.hash hashcat-0.36/examples/A0.M0.word
Initializing with 8 threads and 32mb segment-size...

NOTE: press enter for status-screen

Added hashes from file hashcat-0.36/examples/A0.M0.hash: 102 (1 salts)
e33631b1348f878ef520085ac1da73d7:%O$.#IrVwhp[j@8q2S$,S0{%8+Z,(|>
a3749527010794b356b844e88cbc4b99:= Be|7O0UZtW`nKNXD?(F|$(w*0J+lv)>LYGEW0>+f&T8>g:
78f61d7fafb50ab93345ded65c0dbbc8:'9;E4TE$G',5xN4D`(d?$FVQX=-Rz2guHK{+U<%
dfc9ed604bce8d02fe31d230e762320a:"20]%:^}DZ"l24M;(Qz$W!0=0ECW{%n[I4W}YadPtj ).zo
b19cdd8f0b55cb888c97ec77ab1ad402:rQ-Vv2&kog_p+CMyI;/VFDD.|MQe0/eV6bi9i^"yWP('iT
73ff3971460e3d4490049086207e4a1e:$<-+1Jz?dPL)[;)mg<so'R 5Ri0':.m^Rfv?,x[G
7718d0884dcdd356c77de6ff4c0ad58e:>RHc
15a221074f90b33273146792af9a843b:19s*[J?rKQ6P}
c2e1149440e5340a2a695249247fd435:2.J=982!qStuXOD{*zj}ozg$)
c09b3ca79ea005f7742ade4ac7a9d5c7:G5oZQFPcy)
ac8d9b015966753efc68563f4028706b:G^ExIbyKScFv G9$}CMP
336d5ebc5436534e61d16e63ddfca327:-
168382d511970885b21d9f65a88cde34:L|4BuCOMaZj&7nTGO|=OY7SJHW0;(:I!h6>
add0155c67f5c30b92369d36a6f0cc72:rrF@Vcr]BBlMgm|n`/fwt
f186217753c37b9b9f958d906208506e:O
04841ffd6f854bd06947c8dcfcdb7589:CR[9[Jks>^w/}A;>jsH<
f9d68a3ef957245f752783b73959c2db:,n`eP,LxFL&T3J:paxxR1
06d714a833cbffc22f0fd686699e4b3a:vs\/u|
bc82d57d2a481df9a479346159beef8e:'>+S:T[1\*k>4uX@:2o}hNI9FtzeF\PST*`^GB6CK9\at+Q`G
6d15416d179f0444989e5a6d8193c5ad:&yg0#<RiAR:OF@M@K5^IU\
465133fae5a994afb03c7158260b2e8d:kCQArZz)It
69569b09854879714f024b328dfbf855:)NH&1=G>3:7,!\x8dxM*6gp}*pyf7t
e57725cb60bb641d5160af838f85b59d:y/%,%o^%V}_XJ{;|R";&Jb]*P|y1XoRiXAigC*LB
a5585a8a28488ceafc06e8022779b8bc:K0KIqUx++p=[ Ne\Xhzhfn1E9rjQl?DxAOse^Q=6kbkB'Oy-ESJ|x,
4ab3a9e1c41231114c9d68ea9fc2c7cc:F%{qBL.b9-qZ%#1l.iihek7.huBGbj'_
a55fd320997b2fcfec48edcfbbebfd06:+M1/wp{m/x/ir:Wm@n|htEJ4
0fb840d31afc62afd24b46b12e5de175:EBRtB}NuW-FWu[P${,iS\g`G24zaipl.\9*r?FP=GadrLU=ojR
aa0b039e18dabfd86805449540ff71d1:GZx
c8b5a5c7de28b9033f2e9e892e319a59:Krb=
924d39ab5abba21ac8d0b0c395ad22d0:-HP cS$lsk$)l!Q@?$h:
c562607189d77eb9dfb707464c1e7b0b:LT
88e00bb3096cf71d04a880f636f7e199:+/_W_)0}}(QZ,
62302f17d1566b2fafa630b1c6d6c70b:qnNu0##eXECwoy.Y2
6aea70bd1ed869953cee55f39b103666:(xvm.=ML,IgUkmc q2
a3bd889c3189cfd1435ec38920eb2b68:q)Hgwwe&(zO]s;i`1pC.F644&^{un..}}62D'E.
7b9a47a0714a5260f730959b590cb930:&X 1)$>f^G+mVKxi]{YKL/ttVrq1?|":S!(\qTD
ed69416d83eebf6884ce6e92b467b620:K!VvC\8tR@EPGIY3xM'w|j_U,BdVf!>?
b483c7346ac9fabdb9d40c593df2a14a:#inS2hbk`
8f02e6f714eaaee7d96f6d16525d237c:9e%@oo7MxX_sSI%MY{7e93^aWz:,q
09ae3f58a6f25b12898a4ad933344ba9:B3Hvov$C'OY8>.epecR
b90daad90010ec7bcb239a0ac8d5909e::\6I}wvHDP+[^
074bbfc88bb2cbdf793f252b556dc371:[ri@o.P7^{}RQdl)e|'7j :v;xm||EvsIVk2A) zQ|Ea{8
65b6907db2010d0dbba32949672ade9b:m@V.+=/qr5 c=6|![?in]GMyZ>1@TQkq"aaO1g^+'ZFl`VUK@2
a187f54a3d77d8e7536c4551ce3f9736:1N*Z;(#EI13|awQi8lOx"
a5f3c6a11b03839d46af9fb43c97c188:K
b0422c45899f1c9f133e3ec18bb39ff6:c&8k;kyO
26de1e6f7fd926c0a1fb9876ba4b95ee: )jJE,<_eeX0AzUsPb2;(D["ZM!#LDq8Ms>NyAR|
43bfa82067cf23340aeaad883c71d5c1:6$tFL|N%Cp<A<>lD%
c21308f3c5b637eb27924ef97ee4e256:jrQ(A/tR?8O3l0tHo!O-%-@kXtBy$(9aPavR.vY]7YgENMzF7v^
acb838e2f85962c378ad32f323f07aff:h%z'wwlT0k[TDK\;t=5u{j[z?Jupdy2"%K#*iT
b4efdf3a5f9d3978b3172833a197226d:Bv[yv*sd0JTo^)E{7o.sN3W)6L|[\3
819dcec0c47b4dde9d27a686ae5d2a33:%E_oWS"gV\.qB0{kKjv-Iu|=J*"]f@MY0Dy{B`^kc+<%`
3c9547cb69b13cb32d64eff23a622c3e:[6+U5()pD/9pM
cffc4c9561f6125460048112ec34ff04:XYs)}NG9OR9(V4mST32NL;P8D#Z!g}*^e-
17402b43111bd26ff39eab68730ac678:pl9(54RUGZ?hQg;Fi[25,[$<4
83878c91171338902e0fe0fb97a8c47aTongue
8f0f1cd4790cf5ab7ba473a1e968201f:8+`I
d3e6c8412ab76c9d24276d762a3af5fd:y?'c^+?'rZ1DJbO<vim`L*/MYibPQRKmb]3Y!!eZ
d32ce18738d408462d4676bd15fc4c10:Zd`kQ5>7{
3e9b552241058188391217caa59898ae:b<<G}}d
0d3ffe5be577c7f20e16f84774b74851:l|3!AL,`px2} ro0D9/U<C<yo
a0e1ce47ea7e3fcf415cc0af3c11fdf5:+Ho&!b
ee66a2e6662a74cff2a5ba9a2055d994:*BKLHjHz%mk)Ah]
651e96f9b94e1a3a117eade5e226bd1e:y[N"%e?U{<k[`x<TlG U6Z
58592fe1604f7aeaf127157a99e137bf:QY5H9$9rgsJFLw'Z,`aoniXDT.>J
3a78e5c478cd929e4cabc9e8019ac89c:!Z+i!zNm+S=Ha?/'E
8cff2b58e38df442e3363a31b3aadd11:0@s8'U-4EI9J1/\z2<R
d1457b72c3fb323a2671125aef3eab5d:?
def3dcaab538f1dc42f5f3ec3faf8fc3:2C|3;|p
fca7cb89437fab5d7641931693e60202:B?\G"ouY=F8CRP
9e9881e38fe259ddb0c33a07f9939471:0|mrv qzB lti=a6-pt|@ov(8{y@-Y
13ad68f3d80437113ed4426495bb0e2e:AqR._X+]()Uje@@D=j
3e5dbd5ae3ce51eebd336b54010df7e8:<clDsc
01460379a5b369cda04198a4d349fd36:?Z:c
2a385374677f86bd8969fb5d00a93f6b:g0$mUY-UTQKf8YhSqZ,hZ:IX%BF|"E;'MsHt4'V($0x
56f875bd495305db42be8457477a8831:NBy;OFpqiGn(f)xL|4Wd=NObkf
d4ab9f13c3b66720211ef40deb3d3df9:4M2qx&BqY}&qmxjsl{ uCcH0SkE+
89a1e72161fbede4a9d79c08955a30e2:v* uP|D
26d9cb67c8dad29e54c7961137f8077a:snc`CF[?bVyy4V8)Kw6dWj?b[4Zml
1977fb03158fbd925d6c232cffc1cea2:4d!Wr#68*L
7475e3dd52afb6a8b2e00f3482111008:K6TfXc(w5.XmRy/BzuV[MjbMI
b6ddc0a05b31b438888a9886f5da373a:u.q\EVA*Z(|\s7!^cJ5A]|T{Tf[;zoq5;;X8HxaUfz!nQe,jr
3c8525fc6fcff0dd5720d51efb1f177b:G%k
e5ad218be72f77274a3ba814d32e33bc:xAoRq},[MieH=qW;j"$6Iiazejz%)9Qq*5uZ3}"a6hV
fea2442c8fca322a6a09c72adf579cbc:FLVqF#X53i'?x#yPe%Z,y`}S-pB.`d1(i7BS(P#$
c9ae48033df582a023066e0f9354ff40:#aU{T.Ygz'-
668eb06eddcf2ed9100156c085e930d4:B/5wENJ#n}V|'2EH1si#/,\bOgKDGCW
283f5c47b4bb5b66e74b926c0d1e26a2:`$mCy0HCq$%yuDOO/{qi/GFoyQC[}(jwrZN?.\n}$
f88a619eddbfae9dbe498a762d359b62:$0%u[dhu?#ey`x<x(
409df46291b5e602123cd6957615b4c3::lYHk'V,Z0
c120503b98c8f34edda4921e015a0cd1:+&x7 UJTX{2l\T)k_xlhnnL4=vx8r7O>=-2Do\d(i2}5!TRH04l
7c0eecebff0ad06fd89e25210e70a175:jt;>lt=)BfXOJ= mHPO&P(sUqrM6SJSa!M!"ItY
5733cc0594f5a021065e7df55019a799:{m[21(yB4By$Dy2CI-'2"mte
9ea4f0049593b5fcdf3766742c5eb837:_aPDci3'WX^:NJ%LP3O\I!Rr7-t.y(a#
4381fcb364f3f744cae76c0c28fd640e:GJX-'y)cZD\X-0/Jjh {Q71G<O%|It%)SgMUtOEH@
b117432ceaf165f23834c674a65603db:0r7Xbm|!)|wE3o%=-Z!-nCN4#jkyUP5R7vh4nNdFD
1550b639bc9a6c456f1fe856992a4276Smilev<$-^t*tY}OGi-jG)U\Fvh3Ag^Ihd+0aa&hgxkzB^it\|u
a794cbf461c1aa1a2b13001e7ff400ea:CZF@x?Jq PU1wp:w%RUkJn[{;OI_m'VQ
c64125ba39d3b7a9cd88b27e1649ffa9:qy?`pqb
54b292cf928775b83c16b77a150a3c1c:bj?<lloB.>Qt@3m4|@c9oj(GoPa:00p;a9'-\`Nu?#M
dc370ed85a874916fcf3a1b9345569a8:ava3<3NB8(@G@Sl>Y;r>PV]c:%C8;NIyo@.6n$=is,=\g;PDyh:8
9f7352a74f5260caa85d4fbc048d27f9:%XrcF6h%<kp3.I$'}yvXFBXo^\a=yZ,jljlH%V,)W_v8XwZlD?8W$
All hashes have been recovered


##################oclhc+##############
# oclHashcat+-0.03/oclHashcat+64.bin -m 0 hashcat-0.36/examples/A0.M0.hash hashcat-0.36/examples/A0.M0.word
oclHashcat+ v0.03 starting...

Hashes: 102
Bitmaps: 10 bits, 1024 entries, 0x000003ff mask, 4096 bytes
Rules: 1
Platform: ATI compatible platform found
Device #1: Cayman, 1024MB, 0Mhz, 24MCU
Device #2: Cayman, 1024MB, 0Mhz, 24MCU
NOTE: gpu-accel auto-adjusted to: 8
Device #1: Kernel oclHashcat+-0.03/kernels/4098/m0000.Cayman.64.kernel (1134240 bytes)
Device #2: Kernel oclHashcat+-0.03/kernels/4098/m0000.Cayman.64.kernel (1134240 bytes)

Starting attack in wordlist_mode...

336d5ebc5436534e61d16e63ddfca327:-
d1457b72c3fb323a2671125aef3eab5d:?
f186217753c37b9b9f958d906208506e:O
83878c91171338902e0fe0fb97a8c47aTongue
a5f3c6a11b03839d46af9fb43c97c188:K
c562607189d77eb9dfb707464c1e7b0b:LT
3c8525fc6fcff0dd5720d51efb1f177b:G%k
aa0b039e18dabfd86805449540ff71d1:GZx
8f0f1cd4790cf5ab7ba473a1e968201f:8+`I
01460379a5b369cda04198a4d349fd36:?Z:c
c8b5a5c7de28b9033f2e9e892e319a59:Krb=
7718d0884dcdd356c77de6ff4c0ad58e:>RHc
06d714a833cbffc22f0fd686699e4b3a:vs\/u|
3e5dbd5ae3ce51eebd336b54010df7e8:<clDsc
a0e1ce47ea7e3fcf415cc0af3c11fdf5:+Ho&!b
c64125ba39d3b7a9cd88b27e1649ffa9:qy?`pqb
3e9b552241058188391217caa59898ae:b<<G}}d
def3dcaab538f1dc42f5f3ec3faf8fc3:2C|3;|p
89a1e72161fbede4a9d79c08955a30e2:v* uP|D
b0422c45899f1c9f133e3ec18bb39ff6:c&8k;kyO
d32ce18738d408462d4676bd15fc4c10:Zd`kQ5>7{
b483c7346ac9fabdb9d40c593df2a14a:#inS2hbk`
465133fae5a994afb03c7158260b2e8d:kCQArZz)It
1977fb03158fbd925d6c232cffc1cea2:4d!Wr#68*L
409df46291b5e602123cd6957615b4c3::lYHk'V,Z0
c09b3ca79ea005f7742ade4ac7a9d5c7:G5oZQFPcy)
c9ae48033df582a023066e0f9354ff40:#aU{T.Ygz'-
15a221074f90b33273146792af9a843b:19s*[J?rKQ6P}
3c9547cb69b13cb32d64eff23a622c3e:[6+U5()pD/9pM
88e00bb3096cf71d04a880f636f7e199:+/_W_)0}}(QZ,
b90daad90010ec7bcb239a0ac8d5909e::\6I}wvHDP+[^
fca7cb89437fab5d7641931693e60202:B?\G"ouY=F8CRP
ee66a2e6662a74cff2a5ba9a2055d994:*BKLHjHz%mk)Ah]

Processed 93 words in 105.63ms = 0.88k/s

Started: Fri Feb 25 09:10:07 2011
Stopped: Fri Feb 25 09:10:08 2011
Find all posts by this user
02-25-2011, 12:42 PM
Post: #5
RE: oclhc+ bug?
everything is fine! not a bug. both oclhc and oclhc+ just can not crack hashes with plain of length >= 16
Visit this user's website Find all posts by this user
02-25-2011, 01:15 PM
Post: #6
RE: oclhc+ bug?
(02-25-2011 12:42 PM)atom Wrote:  everything is fine! not a bug. both oclhc and oclhc+ just can not crack hashes with plain of length >= 16

Thanks I see. However as regards the dcc credentials, the plaintexts are definitely all 6 chars long, automatically generated win server 2003 RC2 passwords with complexity requirement set, so they have at least one char of 3 of the 4 charsets ?u ?l ?d ?s., like this:
qo5%hT
12mNoV
etc.

Best J
Find all posts by this user
02-26-2011, 05:51 AM
Post: #7
RE: oclhc+ bug?
(02-25-2011 01:15 PM)acubens Wrote:  
(02-25-2011 12:42 PM)atom Wrote:  everything is fine! not a bug. both oclhc and oclhc+ just can not crack hashes with plain of length >= 16

Thanks I see. However as regards the dcc credentials, the plaintexts are definitely all 6 chars long, automatically generated win server 2003 RC2 passwords with complexity requirement set, so they have at least one char of 3 of the 4 charsets ?u ?l ?d ?s., like this:
qo5%hT
12mNoV
etc.

Best J
With DCC, you need also the username. It's been reported that the maximum length for the username that it takes is currently 8 so check if those are the ones that are not cracked by OCLhashcat+.
Find all posts by this user
02-28-2011, 06:51 PM
Post: #8
RE: oclhc+ bug?
(02-26-2011 05:51 AM)mastercracker Wrote:  With DCC, you need also the username. It's been reported that the maximum length for the username that it takes is currently 8 so check if those are the ones that are not cracked by OCLhashcat+.

Thanks for the hint, but that's not the problem. dcc is fed into oclhc in the format hash:salt which means that they are rejected initially (by throwing a line length exception, see my previous thread). In the hasfile, salts > 8 have been filtered (grep -r ":.\{0,8\}$") ...

I'll try with some sample data so it can be replicated.

Best
J
Find all posts by this user