Hi ya all,
my name is arlequim and i'm new on this board, this is my first message
Lets waste no time: i have generated a working license for my FlexLM 8.1b application with all the features enabled and it was really easy because i have found out all the signs with "runtime debugging" (the fake serial compared byte to byte with the one generated by the vendor daemon). But i'm not happy, i would like to make my lmcrypt.exe, so i have located the l_sg() procedure @ 427487 , then l_n36_buff @ 4012EB and calcseed gave me Seed1=1AB74168 and Seed2=58CAD933. With lmkg i have calculed VK keys for FlexLM behavior 8.1 (lm_code.h parameters...) but when i use lmcrypt on my faked license it gave me an incorrect sign (and i know it should be 7CC8EDA3DC177037). There is something of strange, i dont know ... i thing the ecryption seeds are corrects ... can anyone help me? You will find all the necessary stuff here:
h**p://w*w.megaupload.c*m/?d=6E7D1HAC.
Thanks in advance and best regards

Seeds are not difficult to revert from call l_n36_buff, but difficulty is how to generate special VD HOSTID format license.
Have you checked that your client must use this special license with CLCODE=*** in server line?

I generate normal license with HOSTID=ANY, daemon accepts that. If your client doesn't accept that, you must define that in your SDK to generate that kind of license.

Good luck!

Arlequim

I tried to take a look at your target, but when I follow the h**p://w*w.megaupload.c*m/?d=6E7D1HAC link you posted
I end up on a juicy "sweet and spicy single" page.
nothing wrong with that, but a different target!

Can you upload again, or upload to rapidshare

Regards

the OPEN MIND software use VENDOR hostid.

Hi NoFlexlm,
thanks for your fastest reply
Uhmmm... i think you right, i have generated the CLCODE parameter in license.dat and patched this name inside the vendor daemon (the original was crypted, nice, not? ) @ this point the only way to follow is to modify the the SDK source, but before i want to study better the algorithm inside the lc_checkout of this damned vd. Just one note: many times great teams like LND and Shooters have cracked this appz (T-h-i-n-k-d-e-s-i-g-n) but their releases dont work because the patched inside the vd are wrong. My cr4ck worx 100% but to complete this marvellous target a nice lmcrypt will be appreciated...

Hello RCER,
thanks for your interest on this target, i think its study will be nice for you
at this moment the previous link is still working, anyway i have uploaded the zip here: http://w*w.filefactory.c*m/file/afh515e/n/FlexLM_thinkflx_zip
Ok, if you find out something let me know.
Have a nice day

Hi FoxB,
you right, this appz is included in OPENMIND products but the main release is T.h.i.n.k.3 T.h.i.n.k.d.e.s.i.g.n. nice and powerful cad. Feel free to visit its web site t^h^i^n^k^3.com (remove the ^ chars from here...) Let me know what do you think.

ok guys, i confirm my seeds are correct

the flexlm license generator and eutron dongle emulator it all for this software
and me

Yes, your seeds are right. It is same as my reverted.

now the problem is:
how is calculed the sign 16 hex chars?
that means:
how to modify the lmgr.lib sources in sdk???

anyway the first step (recover the right seeds) is done ...

Unbelievable, the lmcrypt created with correct seeds dont stop on l_string_key !!! At this moment i have no idea this target is really an hardest work!

@Arlequim: this "CLCODE=10021976" is valid host for you?

In "original" code the CLCODE was encrypted inside thinkflx (=BOLOGNA0), so i have replaced it and deleted the algo. But at this moment i think the only solution is modify l_string_key in sdk sources, right??? Btw all tuts are for conventional flexlm targets, but there are a lot of VD customized like this.
In other words this appz (with CLCODE=10021976) is fully working on my pc, but the "real cr4cking team" like LND in previous releases of this software make use of lmcrypt (as you can see by "\" char to formatting lines in license.dat). Any idea?

Look nether, the CLCODE doesnt matter, thinkd3sign is fully working here...

Yes, it should work without CLCODE=*** as I said to you.