Hi. I saw a new release appear today, and I pulled it down to see if some residual problems in port forwarding from openssh-1.2.3 were fixed. The configure script had trouble finding the openssl package, even though I had it installed in the standard place (/usr/local/ssl). I investigated and found that the failed test compile was caused by two nonfatal compiler warnings that memset and RAND_add were being implicitly defined. As a hackaround I inserted #include <string.h> #include <openssl/rand.h> just before #include <openssl/rsa.h> #include <openssl/bn.h> #include <openssl/sha.h> in the test program. configure succeeded, and so did the subsequent compile. I'm rather unfamiliar with the conventions for configure scripts, so I don't know if this is the right way to fix the problem. Phil
On Sun, Apr 30, 2000 at 12:20:23PM -0700, Phil Karn wrote:> Hi. I saw a new release appear today, and I pulled it down to see if > some residual problems in port forwarding from openssh-1.2.3 were > fixed.what problems are you seeing? -markus
On Sun, 30 Apr 2000, Phil Karn wrote:> Hi. I saw a new release appear today, and I pulled it down to see if > some residual problems in port forwarding from openssh-1.2.3 were > fixed. > > The configure script had trouble finding the openssl package, even though > I had it installed in the standard place (/usr/local/ssl). I investigated > and found that the failed test compile was caused by two nonfatal compiler > warnings that memset and RAND_add were being implicitly defined. > > As a hackaround I inserted > > #include <string.h> > #include <openssl/rand.h>That's perfect - applied. Thanks, Damien Miller -- | "Bombay is 250ms from New York in the new world order" - Alan Cox | Damien Miller - http://www.mindrot.org/ | Email: djm at mindrot.org (home) -or- djm at ibs.com.au (work)