Malwarebytes Offers Pirates Its Premium Antimalware Product For Free 111
An anonymous reader writes: If you have a cracked or pirated version of Malwarebytes Anti-Malware (MBAM) product the company has debuted an Amnesty program for you. Venturebeat reports: "If you pirated Malwarebytes Anti-Malware, purchased a counterfeit version of the software, or are having problems with your key in general, the company is offering a free replacement key." CEO Marcin Kleczynski explained the program and his statement reads in part: "When I started Malwarebytes, I absolutely had no idea how successful we would be today. I am extremely grateful for all of the support from everyone and how fast we’ve grown. That being said, I picked a very insecure license key algorithm and as such, generating a pirated key was, and is, very simple.
The problem with pirated keys is that they may collide with a legitimate key just by the sheer numbers. For example, Larry may generate a pirated key that matches the exact key that I already bought. Yes, this is silly, and yes, this is literally the first thing a professional software company thinks of when building license key generation, but when you think you’re building a product for just a few people you don’t hash out these details.
Now we’ve grown up, and we’ve got a new licensing system that we’ve rolled out in stages. The only problem is that we have millions of users that we’ve sold keys to, or a reseller has sold keys to, or we’ve given out keys to without keeping track. It is a mess, and you as a consumer have every right to be upset.
The problem with pirated keys is that they may collide with a legitimate key just by the sheer numbers. For example, Larry may generate a pirated key that matches the exact key that I already bought. Yes, this is silly, and yes, this is literally the first thing a professional software company thinks of when building license key generation, but when you think you’re building a product for just a few people you don’t hash out these details.
Now we’ve grown up, and we’ve got a new licensing system that we’ve rolled out in stages. The only problem is that we have millions of users that we’ve sold keys to, or a reseller has sold keys to, or we’ve given out keys to without keeping track. It is a mess, and you as a consumer have every right to be upset.
Hardware Locking (Score:1)
Re: (Score:3, Informative)
Burned-in MACs? My nForce 2 motherboard's NIC (the nVidia one, not the Realtek one) has a MAC that's user-definable in BIOS.
Re: (Score:1)
Re: (Score:2, Informative)
Re: (Score:1)
There is no "Windows CLI". I even told you that when I said "(Hint - there is no such thing.)".
Windows is a GUI. Windows does not have a CLI. The CLI you are referring to is a faked, extended DOS environment (to various degrees of fakeness depending on your version and bitness of Windows, cmd.exe vs. command.com, etc. ).
It is not Windows. Windows has PowerShell now, but it's not a core part of the OS.
In short, Windows does NOT have a true CLI.
Re: (Score:1)
Re: (Score:1)
There's no such thing as an interface where you can type commands? There are two available (Command Prompt and Powershell). Whether they are any good or not doesn't change their name.
Re: (Score:1)
command.com and cmd.exe are different, both are NOT DOS, and both are NOT Windows.
PowerShell isn't a core part of Windows (yet) and doesn't have anywhere near the coverage necessary to be a true CLI.
These things may be CLIs, but they are not "Windows CLI". There is no "Windows CLI". idbeholda has no idea what he's doing. He seems to think running pulling some strings from some cmd.exe commands let him create strong, hardware-locking DRM.
Re: (Score:2)
The only thing required to make either one a "true" CLI is that there be a command line that lets you interface with things. It does not have to be a core piece of the OS to fit the definition. Otherwise, you could say that Linux distros don't have a true GUI.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Many MACs are adjustable in the drivers.
And yet some are not and are hardcoded. I.e. The Surface Pro series has a MAC that isn't adjustable. The registry hacks don't work because the registry keys don't exist and if you use some software to spoof the MAC you end up in a BSOD loop.
Re: (Score:2)
Or what happends i
Re: (Score:2)
As for an API wrapper, considering that most of the code is
Re: (Score:2)
Re: (Score:2)
The only reason they didn't implement the new system was because they were "afraid they would s
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
If total overhead increase of 200KB for compiled application size, and ~3-5MB memory overhead for non-invasive DRM is a joke, then yes. But not as much as MS extending support until 2024 to allow for the "migration to .NET". At that point, I'll have moved onto other things.
Hopefully you move onto something you understand.
Do you REALLY think you're the first person to think they've got good DRM?
Re: (Score:2)
Re:Hardware Locking (Score:5, Informative)
If Dir("gethwi.bat") "" Then Kill "gethwi.bat"
Open "gethwi.bat" For Append As #1
Print #1, "w32tm
Print #1, "systeminfo >gsys.dat"
Print #1, "getmac >gmac.dat"
Print #1, "exit"
Close #1
Shell "gethwi.bat", vbHide
You use this information to generate an ID. But you don't even hash it with a one way hash, which means it's possible to forge a reply to give an desired result. A good one way hash would at least make that impossible. It is also not scaling very well - you will need a lot of support for pissed customers who changed parts of their computer or changed timezone.
Furthermore, you do no authentication of the answer from the server. Anyone can send the response, and be accepted. You do not have any security. It would be trivial either remove your DRM by jumping over it, or supplying the very wrong values. A race condition would also work - overwriting the gsys.dat, gtime.dat, gmac.dat before your program reads it. Or simply replacing the code snippet above with a batch file which state echo "Desired values..." > gsys.dat.
So take an evening, think about how you can bypass your system. Try my suggestions. Fire up an debugger, and have a look at the software.
Re: (Score:2)
QED
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
If total overhead increase of 200KB for compiled application size, and ~3-5MB memory overhead for non-invasive DRM is a joke, then yes. But not as much as MS extending support until 2024 to allow for the "migration to .NET". At that point, I'll have moved onto other things.
Hopefully you move onto something you understand. Do you REALLY think you're the first person to think they've got good DRM?
He's been repeatedly asked for an executable we can have a bash at, and he's refused (apparently it's too much work). I've seen this on usenet waaaay too many times in the 90's. Some new aspiring unsung-encrypting-genius will pop up on comp.programming (or similar) and boast about their encryption algorithm without giving any details about it. Suffice to say someone usually managed to decode their ciphertext within a few hours.
This appears to be more of the same - at least the usenet newbies had the grace t
Re: (Score:2)
Re: (Score:2)
I can read it just fine - but it does appear to be designed for much lower DPI screens (1024x768 @ 17"). So the design is probably over a decade old.
Do you know how zoom works?
Re: (Score:1)
Re: (Score:2)
Can I move my license over to new hardware without having to rely on the software vendor's cooperation?
Re: (Score:2)
Also, how well does your system stand up to NOP?
Re:Hardware Locking (BIN SPAM) (Score:1)
It's pretty easy to lock down these things via hardware.
In fact,
cough (me think he protest overly so) "to be frank", "to be honest" (trust me, that warm feeling is not me pissing in your pocket, in fact...)
I have a working model that's (theoretically) infinitely scale-able
So much for the fact. The "theoretically" taketh away the impossible "infinitely"....
on any given server, ignoring file number restraints from the hosting provider.
tl;dr? rhetoric + sophism + bullshit = pure_weasel
http://www.tot-ltd.con/WMSDK.h... [tot-ltd.con]
FTFY
Now stop spamming and fuck off back to Intertubes Worriers where you belong. Surely they need more of your fake malware products more than /. readers? (Erick - that is you isn't it? Your self-promotion is almost as good
How stupid could someone be? (Score:2)
>> cracked or pirated version of Malwarebytes Anti-Malware
Really? Could anyone on SlashDot really be this dumb?
Re: (Score:2)
On second thought, it looks like the AV company is staffed with idiots.
>> keys is that they may collide with a legitimate key just by the sheer numbers...when you think you’re building a product for just a few people you don’t hash out these details...
C'mon guys. Your wrote your own clue in the summary. (Starts with "h" rhymes with "trash"...)
Re: (Score:3)
Hash collisions happen.
The real solution is to NOT use a generation algorithm for keys. Generate strings, then approve only those you actually sell and distribute.
Software installation/runtime checks locally against the generation algorithm, allowing for offline installations, bundled installers, old version installs, use in 50 years after all the servers are gone, etc.
Updates ask for your key and the server decides if it's valid (an approved string that hasn't been used by thousands of PCs across the net
Re: How stupid could someone be? (Score:3)
To expand on this... you should also generate an "Installation ID" upon validation, stored server and client side along with the key.
This prevents users from trying to activate the key on more than one system, and allows you to offer controlled multi-system installs if you so choose.
On update you validate both the key, and the installation ID.
In the event a user needs to move the software to another install, you can contact the licensing dept and revoke the previous installation ID.
Re: How stupid could someone be? (Score:2)
To further expand on this... keep talking, meanwhile as a client I'll be looking for software with none of that crap.
Re: (Score:3)
Please don't try to limit the number of installs. It breaks horribly when you try to do a re-install, or move to a new PC, or run in a VM. The nature of this software is that techs will often install it on customer's PCs, clean them and then remove it.
Install counters are evil.
Re: How stupid could someone be? (Score:2)
Really depends on the nature of the software, I guess. For Malwarebytes it probably isn't the best idea, but at the same time it could easily de-reg the install ID upon uninstall.
There are various ways to do it. My example was one such way, that is all. There is no one-size-fits-all.
Re: (Score:2)
Hash collisions will eventually happen. I believe Windows XP suffered from it where the sheer number of installations has meant that there was a good chance a keygen will also make a valid key that's already been issued. Sure you are blocking a good chunk of them at the beginning, but eventually a keygen will stumble upon a valid key that you DID issue.
I believe it also
Re: (Score:2)
Generation algorithms for software license keys is fine.
Simple generated keys stop casual sharing of licenses. Nothing stops dedicated hackers.
Why invest time and money in a very expensive license key system when all you're doing is providing the hackers with a more interresting challenge.
The problem here isn't generating keys, it's the relatively high chance of colission; it's badly generating keys.
Re: (Score:2)
On second thought, it looks like the AV company is staffed with idiots.
Yeah that pretty much sums it up. I have a legit copy through work and it gave me the 'blahblahblah ur a pr8' bit the other day.
Re: (Score:1)
Have you read slashdot commentary lately?
It's a great idea (Score:2)
It's a good marketing move - most people just download the free version and scan. Problems fixed so they won't buy it for the bells and whistles - now they'll get lots of people to try the bells and whistles and might retain future revenue.
It's better than them canceling the free version and make it pay only for revenue.
Re: (Score:1)
I have used their service in the past but I never bought their program. I did some debugging and definitions work with a security company as a lark and have free access to their software. Anyhow, it took a minute to find a code online which I entered (use the ID too) and it offered me a free key after that. It was pretty painless. I will give their monitoring service a try on this laptop and see how it works out. I often do not use real-time AV anyhow.
Fair is Fair? (Score:1)
Re: (Score:1)
Pretty much, I have almost no incentive to buy the product if I can pirate it then get it for free
Reddit (Score:1)
Snake oil product vendors (Score:2)
A software company showing respect for customers? (Score:1)
Damned if this isn't a first. I've never needed a licensed version of their software, but the transparency, respect, and benefit of the doubt they are giving users of their software, both paid and otherwise, truly impresses me. As such, I will be happy to purchase a license the next time I need their software.
Only on slashdot.. (Score:2)
..would people shit on someone for acknowledging a problem, admitting fault, and then moving to fix it in a way that benefits not just the consumer, but everyone else too.
Fishy (Score:1)
Re: (Score:2)
Re: (Score:3)
Short answer is no. Long answer is yes.
Some versions of the crypto viruses have the keys released so you can decrypt. Others do not. If you know how to google, you know how to find out what version it is and if it's been released. If you've got cryptolocker it's simple.