Windows 3.11

[Bailey]

This is a long shot but does AxCrypt support Windows for Workgroups?

I’m the sysadmin and we’ve got a number of industrial control systems that run Windows 3.11. To upgrade the hardware would cost us many millions and is out of the question.

We’ve got a new regulatory requirement to use data encryption on the systems and I’m struggling to find a solution. We might have to look at testing hardware-based drive encryption but this may not work as we tried it unsuccessfully on our more modern z/OS mainframes.

I’ve come across PKZIP for MS-DOS but that only supports ZipCrypto and is unacceptable because we need a minimum of 3-DES or one of the approved AES (128, 192, 256) algorithms.

Is there a really old version of AxCrypt that support Windows 3.11?

Thank you.

[SvanteSpectator]

Hello Bailey!

Now that’s an interesting question… I’m sorry, but I think Windows 95 is the earliest version we ever supported. I’m just guessing, but I’d think you could find some old PGP version that might run on 3.11 and it might support 3-DES. Possibly also some early ports of unix crypt, although that might only support DES, not 3-DES (not sure).

Considering the alternatives, and assuming that you might get away with a simple command line utility, I’d try to setup a development environment for 3.11 and get some old C sources for crypt for example from minix or some early linux distro, and maybe swap the algorithm for AES – there are reference implementations in pure C that should compile. Turbo C was pretty good back in the day ;-)

[Bailey]

Thanks for such a wonderfully quick reply.

I did have a look at your old webpage <axantum.com> and <axondata.se> prior to posting my question but I could only find mention of the 32-bit edition.

Great suggestion about PGP! We’ll reach out to Symantec to see if they have any licensable copies of PGP for Windows 3.11 but very few vendors support it any more. I remember that at one time it was embargoed software so we’ll need to make sure we don’t get an export [weakened] version.

I knew it was a long-shot and I really wish I didn’t have to support such old architecture but with systems cost $40m+ to replace it looks like we’re going to have to develop a 3DES solution ourselves. I vaguely recall Borland Turbo C but I’m not familiar with it and we’ll probably end up hiring somebody who remembers the older languages.

We’ve only just managed to upgrade from 5.25″ floppies because of compatibility problems with the “newer” 3.5″disk – I’m not joking :-(

 

[Bailey]

The oldest version I could find was AxCrypt 1.1 and that’s for Windows 95.

I’ve managed to locate <AxCrypt-Src-1.1.zip> which is the source material so I take a look at that and see if there’s anything we can use.

Kind Regards

[SvanteSpectator]

Hello,

Re PGP, check out the PGPi (I think it was called) version. During a period, they exported it on paper and then scanned it in europe, because exporting a book was ok even if printed in only one copy but exporting encryption code was not ;-). This was not a weakened version, but I really don’t remember exactly what years this was.

If you’re interested to hire someone to build something, send me a PM. I don’t think I have any media left from that time, but it appears it might be possible run Windows 3.x in Virtualbox or VMWare. It ought to be possible to get hold of a Turbo C copy somehow also.

[Bailey]

Thank you again for your suggestions.

Amazingly a colleague of mine has sourced PGPfreeware 6.5.8 which supports CAST, IDEA and 3DES. It also provides command line functionality.

A local university (we give them free use of our scientific instruments) had a copy in their computing faculty.

We’re going to give it a try but if that fails then I’ll drop you a PM. We have original installation media for Windows 3.11 and MS-DOS 6.22 and we probably could source Turbo C (if we don’t have it already somewhere).

[Author]

Posts