hashcat Forum

This is not related to my ocl-hashcat-plus v0.09 beta report. This is for ocl-hashcat-plus v0.08 stable.

I was running through my usual lists and rules files on a .hccap file that works fine with ocl-hashcat-plus v0.07 with ocl-hashcat-plus v0.08 when I noticed that ocl-hashcat-plus v0.08 froze on me.

Using Win7 Ultimate 32bit ATI HD5870 Card Driver Package Version 8.961 OpenGL Version 6.14 against .hccap

My first rule list ...



My second rule list was suffixing 0-100. (To much to paste.)

After about 180 minutes I noticed that my fan speed had reduced to tick over so I turned the monitor on expecting to see the cracked notification as the time estimate was over 16 hours when I started and as I said this was just about an hour and a half in.

However I am unsure what was happening as I noticed that the WPA had not been cracked but ocl-hashcat-plus had stopped cracking. Well I say "had stopped" I could press "s" and get a status report as if it was still working. The c/s figures were still showing what I would expect to see if it was working full tilt.

I did notice that the Progress and Rejected figures were not changing every time I pressed "s". If ocl-hashcat-plus was rejecting many thousands of passwords due to line length etc, then I would expect to see the Rejected figure to change, but it wasn't doing.

I left it like this for 15 minutes half expecting it to pick up again but it didn't. I continued to press "s" and got the same reply.

I finally stopped ocl-hashcat-plus but I wonder what is wrong ?

Just thought I should mention my temperature was fine, I don't overclock and it usually runs at about 70c - 72c.

Thanks

I have just repeated the test with the same results. Runs for about 180 minutes and then fan speed drops to tick over and the Progress and Rejected figures stay the same however many times I press "s" to check.

I will leave it longer this time to see if it eventually starts again.

Just for my own notes when returning to this I got about 11% progress.

Another detail is that the list is sorted small to large. I don't expect oclhashcat-plus to test passwords larger than 15 but I would expect the Rejected number to change.

I think I am narrowing this down.

I have made a full pass using exactly the same set up as before even the same .hccap. I have run through the list completly without using any rules Which worked fine.

I then used the first rule... below, without the second rule (which suffix's numbers 0-100.)

The first rule list.


Then I made another pass, everything the same except this time I didn't use the first rule I just used the suffix numbers rule on its own, which worked fine.

Rule 2

So... I would suggest the problem is only when the two rules are being used at the same time. I wonder if it is the "E" option as that is new to v0.08 ?

It takes so long to make a pass I will have to report back in the morning, I will set it off tonight using both rules but removing the "E" option.

If that pass succeeds then I think the "E" option is at fault, its works on its own but it doesn't like it when there is something suffixed afterwards. This would make sense as I have never used the "E" option on v0.07 and this is only my second day with v0.08.

Try to reproduce with beta pls

(06-05-2012, 09:58 AM)atom Wrote: [ -> ]Try to reproduce with beta pls

OK Will do ! Is this something known in v0.08 ?

Anyway just to provide a little more info. The test last night froze again, along with my idea that it could be the "E" rule as it wasn't present in the first rule file.

It just seems to be when 2 rule files are used.

I have started the test again with the beta v0.09. I am using the same .hccap, wordlist and rules. Its a shame I have to keep testing on this large list as it can take a few hours to see if it works or not but I dare not change anything else in case it gives a false result.

I have to say, the fan speeds is nearly 100% with exactly the same settings as before when using v0.09 !!!

Will report back as soon as I know anything.

This might be the new retain speed feature. It tries to hold the temperature of your GPU at a given temperature (default is 70c). In case temp is above it raises the fan speed by 1% each second. You can disable it with --gpu-temp-retain 0 or simply raise it.

(06-05-2012, 12:39 PM)atom Wrote: [ -> ]This might be the new retain speed feature. It tries to hold the temperature of your GPU at a given temperature (default is 70c). In case temp is above it raises the fan speed by 1% each second. You can disable it with --gpu-temp-retain 0 or simply raise it.

Yes, I suspect this is so, thank you.

Just an update for you, this last test is still running and it has been going for nearly 6 hours which is much longer than when trying with v0.08.

I am starting to think that v0.09beta has this bug fixed.

I will let you know when it has competed fully.

Thanks.

Well that's it !

It has just completed a full pass (a little earlier than predicted) but it finished properly and all seemed ok.

Well done atom for fixing a problem before I found it !! ha ha !

I wish I had just started to use v0.09b rather than v0.08 in the first place.

So what was it ? Was it a rule thing in the end ?

Anyway thank you very much for v0.09b, works quite a bit quicker than v0.08 on hhcaps....awesome !

(06-05-2012, 12:39 PM)atom Wrote: [ -> ]This might be the new retain speed feature. It tries to hold the temperature of your GPU at a given temperature (default is 70c). In case temp is above it raises the fan speed by 1% each second. You can disable it with --gpu-temp-retain 0 or simply raise it.

Nice to hear this ! Some days ago I gave a suggestion in this direction to mastercracker,
and now there it is (I'm not saying that it was my idea or sugestion, instead, it was good to
see it implemented !)

Thanks for that very important detail, atom. At least we don't have to worry tunning GPU speed funs.