I have a problem with the 0.9.4 version of openSSL on Windows NT (this does
not appear to be a problem on Unix). It looks like a similar problem
mentioned in earlier postings by Steven Reddie (see: [BUG?]
BIO_sock_should_retry() on Win32); one major difference is that we are not
using pthreads in our NT application.
While using SSL_write, it returns a -1 for the number of bytes written. I
tracked down the line generating the error; it occurs during the
get_server_hello process:
ret=readsocket(b->num,out,outl);
(located in bss_sock.c: line 190)
"readsocket" is actually a macro for "recv". A WSAEWOULDBLOCK error is
generated here meaning that the receive operation would block even though
the
socket is marked as nonblocking.
In one of the earlier postings, Steven mentioned he made his openSSL work
temporarily by simply removing the characters "&& 0" from the following line
of
code:
#if defined(WINDOWS) && 0 /* more microsoft stupidity? perhaps not? Ben
4/1/99 */
(located in bss_sock.c: line 347)
This unfortuneately does not work for me--I get the same errors.
I used Purify to see if there are any memory leaks etc. I don't know if this
is the cause, but there were a number of warnings generated before it
reached
this line, occurring mostly around the following sections of code:
Uninitialized Memory Reads:
SHA1Update [sha1dgst.c:167]
SHA1Update [sha1dgst.c:168]
SHA1Update [sha1dgst.c:171]
SHA1Update [sha1dgst.c:140]
SHA1Final [sha1dgst.c:461]
I tried the 0.9.6 version of openSSL, but I still have the same problem.
Does anyone have any suggestions about how I can fix this?
Paul Kudlawiec
Software Engineer
Transnexus
[EMAIL PROTECTED]
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
