[comp.unix.sysv386] crypt in European SCO Unix

rvdp@cs.vu.nl (Ronald van der Pol) (12/13/90)

Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?

--
		Ronald van der Pol  <rvdp@cs.vu.nl>

tim@delluk.uucp (Tim Wright) (12/13/90)

In <8493@star.cs.vu.nl> rvdp@cs.vu.nl (Ronald van der Pol) writes:


>Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?

Well, I don't know what SCO's answer is, but with the rule change on DES
encryption (i.e. it can now be exported) they should be able to ship it.
I would guess that they'll want it in writing that it is now legitimate
before doing so though !

Tim
--
Tim Wright, Dell Computer Corp. (UK) | Email address
Bracknell, Berkshire, RG12 1RW       | Domain: tim@dell.co.uk
Tel: +44-344-860456                  | Uucp: ...!ukc!delluk!tim
"What's the problem? You've got an IQ of six thousand, haven't you?"

jfh@rpp386.cactus.org (John F Haugh II) (12/13/90)

In article <8493@star.cs.vu.nl> rvdp@cs.vu.nl (Ronald van der Pol) writes:
>Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?

It should include "some" form of crypt(), but not the DES form.  That
technology cannot be exported from the US to Europe.  You should get
the version that was posted from Australia if you want DES encryption.
-- 
John F. Haugh II                             UUCP: ...!cs.utexas.edu!rpp386!jfh
Ma Bell: (512) 832-8832                           Domain: jfh@rpp386.cactus.org

ronald@robobar.co.uk (Ronald S H Khoo) (12/14/90)

jfh@rpp386.cactus.org (John F Haugh II) writes:

> >Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?
> 
> It should include "some" form of crypt(), but not the DES form.  That
> technology cannot be exported from the US to Europe.  You should get
> the version that was posted from Australia if you want DES encryption.

Um, apparently we can have DES technology for authentication, but
not for keeping our CV's private from our boxes.  If you need
UNIX crypt(3) for passwd authentication, you can have a castrated
so-called "International" version of the crypt library from support
or by anon uucp.  This version leaves out the encryption stuff, but
leaves crypt(3) intact, as far as I can tell.

SCO don't ship encryption as standard -- their stock control can't
manage to keep the export orders separate, after all, they are a
computer technology compay :-) :-) :-)

You always need to ask for a support disc if you want crypt(1),
but you need to be in the USA to do this.

Yeah, John's right, get the comp.sources.unix stuff for encryption,
but get lng225b from SCO if you need crypt(3) to compile John's shadow
password stuff with :-)
-- 
ronald@robobar.co.uk +44 81 991 1142 (O) +44 71 229 7741 (H)

jfh@rpp386.cactus.org (John F Haugh II) (12/14/90)

In article <1990Dec14.025613.21542@robobar.co.uk> ronald@robobar.co.uk (Ronald S H Khoo) writes:
>Yeah, John's right, get the comp.sources.unix stuff for encryption,
>but get lng225b from SCO if you need crypt(3) to compile John's shadow
>password stuff with :-)

Oh, and speaking of shadow password stuff, time for a shameless plug.

The version which I just posted to alt.sources is the old level of
code.  If you've been hearing about new stuff for the last few months,
it is still in the pipeline.  The most exciting parts are libraries
which allow you to access the security information (passwd, group,
shadow, gshadow) in a database-like fashion with a get/put/commit-like
interface and a complete DBM interface for all of those files.

I will take bug reports for that code level, but expect many of them
to have been replaced with new bugs. ;-)

Also, if someone would freight me a SCO UNIX box, I will try to port
the code there and get rid of that stupid SecureWare pseudo-security
code.  If you need a security bond, escrow account or bank references,
I am sure we can work that part out.

jc@joker.mil.ufl.edu (Jim Castleberry) (12/15/90)

In <8493@star.cs.vu.nl> rvdp@cs.vu.nl (Ronald van der Pol) writes:
>Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?

From SCO's sls catalog of 9/13/90:

SLS #lng190     SCO Xenix/Unix Crypt Support
For SCO UNIX/XENIX             version: 2.2.3 to 3.2.0   machine: n386
   This Support Level Supplement (SLS) provides crypt libraries and editor
   binaries for SCO XENIX and UNIX Operating Systems.  This supplement
   obsoletes lng073.  May be shipped to domestic destinations only.

SLS #lng225     International Crypt SLS
For SCO UNIX Development Syst  version: 3.2.0       machine: U386
   This Support Level Supplement provides the International Crypt Library, and
   may be shipped to customers outside of the Continental United States.

I think lng225 is what you want, but they still make you ask for it.

em@dce.ie (Eamonn McManus) (12/20/90)

rvdp@cs.vu.nl (Ronald van der Pol) writes:
>
>Should the European version of SCO Unix contain crypt(C) and/or crypt(S)?

I'm not sure if I should mention this for fear that it will be "fixed".
However if the export rules are being relaxed presumably it won't have to be.

Although the crypt() library function was removed from libc.a, SCO forgot
to remove it from the shared version libc_s.a.  So you can link a program
(such as our un-C2ed su) that uses crypt() by giving -lc_s on the command
line.

,
Eamonn