09-11-2011, 12:03 AM
When a substitute rule (eg. sa@) would not change anything to a word candidate (eg. "secret" does not have any 'a'), hashcat outputs a weird/unpredictible word, although it should skip the word instead and not output anything.
Using hashcat-0.37, steps to reproduce:
$ cat > /tmp/test.rule <<'EOF'
^1
sa@ ^1
EOF
$ echo secret > /tmp/test.dic
$
$ ./hashcat-cli32.bin --stdout -r /tmp/test.rule /tmp/test.dic
1secret
secrett
$
-> Notice the double 't' ... somethings wrong!
Expected output should be:
$ ./hashcat-cli32.bin --stdout -r /tmp/test.rule /tmp/test.dic
1secret
$
This bug has a direct impact on hashcat performance when using many substitute rules, because the same weird words will be duplicated many many times.
Using hashcat-0.37, steps to reproduce:
$ cat > /tmp/test.rule <<'EOF'
^1
sa@ ^1
EOF
$ echo secret > /tmp/test.dic
$
$ ./hashcat-cli32.bin --stdout -r /tmp/test.rule /tmp/test.dic
1secret
secrett
$
-> Notice the double 't' ... somethings wrong!
Expected output should be:
$ ./hashcat-cli32.bin --stdout -r /tmp/test.rule /tmp/test.dic
1secret
$
This bug has a direct impact on hashcat performance when using many substitute rules, because the same weird words will be duplicated many many times.
btw i have already fixed this in v0.38 beta.