Salz, Rich wrote:
>
> > (a=b) && (a=BN_dup(b));
>
> Oh gak, are we really trying to be clever for its
> own sake?
>
> >But back to the original question. I think it's really
> >Bug and should be handled by BN_dup() internally.
>
> I dunno. I prefer that functions coredump when given
> N
> I tried a = BN_dup(b);
>
> if b is NULL, it causes error. Is this normal or should it just
> duplicate NULL over to a?
It is normal. You can't do anything with uninitialized BIGNUMs.
Arguably OpenSSL should check that the input is valid first, but
that applies to many other functions as well.
otherwise you get busy waiting. (I'm not even sure that s_client
is
correct in this respect, but it's to confusingly written to deeply
look into this issue.)
_
Bodo,
Good point, since as far as most users are concerned s_client.c and
s_server.c are the most referenced
I've been using the IAIK/iSaSiLk classes for some time in an
cross-platform SSL proxy. They work quite well and are the only ones I've
found thus far with support for JDK 1.1.
I don't think they offer source, however. At least, not under the
demo/educational licensing.
On Fri, 30 Jul 1999,
At 11:20 PM 7/29/99 -0400, Steven J Sobol wrote:
>I tried using CA.pl to create a certificate signed by my own CA, and
>when I try to use it with Apache and mod_ssl I get this:
>
>29/Jul/1999 23:17:51] [info] Init: 1st startup round (still not detached)
>[29/Jul/1999 23:17:51] [info] Init: Initi
David,
> Does anyone know of an open public source version of SSL for Java that
> resides outside the US. I am not keen on purchasing SSL from RSA in the US
> and having to apply for the export license and all.
Have a look at http://jcewww.iaik.tu-graz.ac.at/ They have some very
nice tools ther
Mark Borgerding wrote:
>
> Ben Laurie Wrote
> >It's at times like this that I wish they hadn't made && boolean.
> >Wouldn't it be cool to write:
> > a=b && BN_dup(b);
>
> Yup, that would be cool, but you can still do
> a = b && a = BN_dup ( b );
> or more explict, but still tight:
> a = ( b ?
The JCSI classes from Australia are very good. They're open source and
support SSLv2,3,TLSv1, Kerberos, and various and sundry generic crypto.
The classes are all JDK 1.2, which is fine for our use under Windows 9x,
but annoying on the MacOS (JDK 1.1.7 as of the latest MRJ :(
Look at:
http://s
"Salz, Rich" wrote:
>
> > (a=b) && (a=BN_dup(b));
>
> Oh gak, are we really trying to be clever for its
> own sake?
Yes. Is there something wrong with that? Bear in mind that I was
disussing language architecture, not how OpenSSL should work.
> >But back to the original question. I think
Does anyone know of an open public source version of SSL for Java that
resides outside the US. I am not keen on purchasing SSL from RSA in the US
and having to apply for the export license and all.
Any suggestions?
Thanks!
David
-Original Message-
From: David Smith [mailto:[EMAIL PROTEC
> (a=b) && (a=BN_dup(b));
Oh gak, are we really trying to be clever for its
own sake?
>But back to the original question. I think it's really
>Bug and should be handled by BN_dup() internally.
I dunno. I prefer that functions coredump when given
NULL pointers.
/r$
__
Ben Laurie Wrote
>It's at times like this that I wish they hadn't made && boolean.
>Wouldn't it be cool to write:
> a=b && BN_dup(b);
Yup, that would be cool, but you can still do
a = b && a = BN_dup ( b );
or more explict, but still tight:
a = ( b ? BN_dup ( b ) : 0 );
or better yet, we coul
Holger Reif wrote:
>
> > >> I tried a = BN_dup(b);
> > >>
> > >> if b is NULL, it causes error. Is this normal or should it just
> > >> duplicate NULL over to a?
>
> > > It's at times like this that I wish they hadn't made && boolean.
> > > Wouldn't it be cool to write:
> > >
> > >a=b && BN
Announcement: OpenSA version 1.3.6c fixed
OpenSA - the Open Source Server Package
http://www.opensa.de/
The OpenSA project team is pleased to announce the release of version 1.3.6c
of our open source packge for Win32. This new OpenSA version incorporates
several changes and bugfixes to the packag
Julio Sánchez Fernández wrote:
>
> If I do openssl x509 -x509toreq, I need a private key. However,
> function X509_to_X509_REQ allows the pkey argument to be NULL.
>
> Why do I need this? Can I just modify x509.c not to insist in
> finding the private key?
>
The private key is needed to sign
Bodo Moeller wrote:
>
> On Fri, Jul 30, 1999 at 09:49:57AM +0100, Ben Laurie wrote:
>
> >> I tried a = BN_dup(b);
> >>
> >> if b is NULL, it causes error. Is this normal or should it just
> >> duplicate NULL over to a?
>
> > It's at times like this that I wish they hadn't made && boolean.
> >
> >> I tried a = BN_dup(b);
> >>
> >> if b is NULL, it causes error. Is this normal or should it just
> >> duplicate NULL over to a?
> > It's at times like this that I wish they hadn't made && boolean.
> > Wouldn't it be cool to write:
> >
> >a=b && BN_dup(b);
> >
> > ? (Works in Perl, BTW).
If I do openssl x509 -x509toreq, I need a private key. However,
function X509_to_X509_REQ allows the pkey argument to be NULL.
Why do I need this? Can I just modify x509.c not to insist in
finding the private key?
Thanks in advance,
Julio
_
On Fri, Jul 30, 1999 at 08:15:01AM +0200, Holger Reif wrote:
>> What signals a key renegotiation during a SSL session? Do Nav/IE typically
>> renegotiate session keys?
> Yes, in case a client cret is requested later by the server
> (depending on the URL requested) and ibn case of Global
> Serve
On Fri, Jul 30, 1999 at 09:49:57AM +0100, Ben Laurie wrote:
>> I tried a = BN_dup(b);
>>
>> if b is NULL, it causes error. Is this normal or should it just
>> duplicate NULL over to a?
> It's at times like this that I wish they hadn't made && boolean.
> Wouldn't it be cool to write:
>
>a=
Steve Lihn wrote:
>
> I tried a = BN_dup(b);
>
> if b is NULL, it causes error. Is this normal or should it just duplicate NULL over
>to a?
> Can someone fix it if it is a bug!
It's at times like this that I wish they hadn't made && boolean.
Wouldn't it be cool to write:
a=b && BN_dup(b);
On Fri, 30 Jul 1999, Holger Reif wrote:
> Steven J Sobol schrieb:
> >
> > On Thu, Jul 29, 1999 at 05:03:20PM +1000, Damien Miller wrote:
> >
> > > > You can have a look at the mkcert.sh script of mod_ssl. This might
> > > > be a good starter.
> > >
> > > ... or better yet, the CA.pl included wi
22 matches
Mail list logo
