Same hashes different results

[Jogjab]

As for your other questions...

I started using Linux yesterday. So I hope I'm anwsering you're questions correctly:
I'm running:
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=11.10
DISTRIB_CODENAME=oneiric
DISTRIB_DESCRIPTION="Ubuntu 11.10"

And it's cuda

Did some test now but was unable to reproduce this. Are you sure that the charset matches with the passwords, i.e. that all hashes *could* definitely be cracked by the mask? E.g are there no special characters etc.

Please also tell us what hardware do you use (cuda/ocl - card model) and/or which kernel is loaded.

MY TESTS:

Code:

// Description: Generates the MSSQL 2005 hashes starting from a password and salt // Date: 05/22/13 // // License: belongs to the PUBLIC DOMAIN, donated to hashcat, credits MUST go to hashcat // and me for their hard work. Thx // Disclaimer: WE PROVIDE THE PROGRAM “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER // EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE // Furthermore, NO GUARANTEES THAT IT WORKS FOR YOU AND WORKS CORRECTLY // // HOWTO compile: gcc mssql_2005_compute.c -o mssql_2005_compute -lcrypto // Example usage: ./mssql_2005_compute hashcat 18102152 // 0x010018102152f8f28c8499d8ef263c53f8be369d799f931b2fbe #include <string.h> #include <stdio.h> #include <openssl/sha.h> #define DIGEST_LENGTH 20 #define MSSQL2005_IDENTIFIER "0100" #define MAX_PASS_LEN 50 #define MAX_LEN_SALT 8 // 8 octets == 4 hex #define MAX_LEN_UNICODE_SALT 105 #define HEX_BLOCK_SIZE 2 char*HEX_CHARS="0123456789abcdef"; int generate_hash(char digest[DIGEST_LENGTH],char*salt,char*pass) { int i,len_pass,len_unicode_salt; char unicode_salt[MAX_LEN_UNICODE_SALT]; len_pass=strlen(pass); if (len_pass>MAX_PASS_LEN) { printf("[-] ERROR: password too long. EXIT.\n"); return 0; } // Convert pass to uppercase unicode memset(unicode_salt,0,MAX_LEN_UNICODE_SALT); len_unicode_salt=len_pass*2; for (i=0; i<len_pass; i++) { unicode_salt[i*2]=pass[i]; // Unicode } // Salt memcpy(unicode_salt+len_unicode_salt,salt,4); len_unicode_salt+=4; SHA_CTX context; SHA1_Init(&context); SHA1_Update(&context,(unsigned char*)unicode_salt,len_unicode_salt); SHA1_Final(digest, &context); return 0; } void print_hash(char*salt,unsigned char digest[DIGEST_LENGTH]) { int i; printf("0x%s",MSSQL2005_IDENTIFIER); for (i=0; i<sizeof(salt)/sizeof(char); i++) { printf("%02x",salt[i]); } for (i=0; i<DIGEST_LENGTH; i++) { printf("%02x",digest[i]); } printf("\n"); } int parse_salt(char salt[MAX_LEN_SALT],char*input) { int i,j,len_hex=strlen(HEX_CHARS),power,len=strlen(input); char*pos; if (len<MAX_LEN_SALT) { fprintf(stderr,"[-] Error: salt is too short, length of %i expected\n", MAX_LEN_SALT); return 1; } if (len!=MAX_LEN_SALT) { fprintf(stderr,"[!] Warning: salt should be *exactly* of length %i, ", MAX_LEN_SALT); fprintf(stderr,"using only first %i bytes\n",MAX_LEN_SALT); } memset(salt,0,MAX_LEN_SALT+1); // null terminator included for (i=0; i<MAX_LEN_SALT; i+=HEX_BLOCK_SIZE) { for (j=0,power=len_hex; j<HEX_BLOCK_SIZE; j++,power/=len_hex) { pos=strchr(HEX_CHARS,input[i+j]); if (pos==NULL) { fprintf(stderr,"[-] Unexpected character encountered. *Not* a valid"); fprintf(stderr," hex char. EXIT\n"); return 1; } salt[i/HEX_BLOCK_SIZE]+=(pos-HEX_CHARS)*power; } } return 0; } void usage() { printf("--- MSSQL 2005 COMPUTE ---\nUSAGE:\n ./mssql_2005_compute "); printf("<PASS> <HEX_SALT>\n"); } int main(int argc,char**argv) { int i; unsigned char digest[DIGEST_LENGTH]; // COMMAND LINE PARSING if (argc<2) { fprintf(stderr,"[-] Please specify the password. EXIT\n"); usage(); return 1; } if (argc<3) { fprintf(stderr,"[-] Please specify a salt. EXIT\n"); usage(); return 1; } // END COMMAND LINE PARSING char salt[MAX_LEN_SALT+1]; if (!parse_salt(salt,argv[2])) { if (!generate_hash(digest,salt,argv[1])) { // success print_hash(salt,digest); return 0; } else { fprintf(stderr,"[-] Error: password generation was *not* successful"); return 1; } } return 1; }

My salt from example (http://hashcat.net/wiki/doku.php?id=example_hashes ): 18102152

My test file: hash_mssql2005
5 lower case letters as per example above

Code:

$ ./mssql_2005_compute testa 18102152 > hash_mssql2005 $ ./mssql_2005_compute testb 18102152 >> hash_mssql2005 $ ./mssql_2005_compute testc 18102152 >> hash_mssql2005 $ ./mssql_2005_compute hashc 18102152 >> hash_mssql2005

My command line:

Code:

$ cudaHashcat-plus32 -a 3 -m 132 -1 ?l?u?d hash_mssql2005 ?1?1?1?1?1

AND

Code:

$ cudaHashcat-plus32 -a 3 -m 132 -1 ?l hash_mssql2005 ?1?1?1?1?1

The result of my tests are that I always get all hashes cracked and "Status.........: Cracked" in the output (as expected).
Could you try to do the same?