Hi,
4.0.6 claim about "The procedure entry point freeaddrinfo could not be located in the dynamic link library WS2_32.DLL"
4.0.7 and above crashes "...has generated errors".
Windows 2000 is IPv6 enabled by
http://msdn.microsoft.com/downloads/sdks/platform/tpipv6.asp with information from: http://msdn.microsoft.com/downloads/sdks/platform/tpipv6/faq.asp
Is it planned to fix this issue?
BTW: IPv6 enabled 4.0.11 works fine on Linux !
Peter
"Peter Bieringer" [email protected] wrote:
4.0.7 and above crashes "...has generated errors".
[cut]
Is it planned to fix this issue?
Did you try: ftp://ftp.mirt.net/stunnel/stunnel-4.12rc2-installer.exe ? Can you provide some more information about the crash? I'm pretty sure it's possible to get some more information (at least stack backtrace). In fact I don't have any Windows 2000 machine to debug your problem.
BTW: IPv6 enabled 4.0.11 works fine on Linux !
Linux is my main development platform. Even the Windows binaries and installer are built on Linux. 8-)
Best regards, Mike
--On Thursday, September 22, 2005 06:57:15 PM +0200 Michal Trojnara [email protected] wrote:
"Peter Bieringer" [email protected] wrote:
4.0.7 and above crashes "...has generated errors".
[cut]
Is it planned to fix this issue?
Did you try: ftp://ftp.mirt.net/stunnel/stunnel-4.12rc2-installer.exe
Same issue. I've reproduced this on a clean W2K machine, 4.0.11 worked. Install IPv6 support, 4.11 and 4.12rc2 crashes after start.
? Can you provide some more information about the crash?
I start stunnel on command line and the "..has generated errors.." window rises up after IPv6 is activated.
I'm pretty sure it's possible to get some more information (at least stack backtrace). In fact I don't have any Windows 2000 machine to debug your problem.
How to debug on a Win2K machine? Do you know any tools? I have only found a DrWatson output in event viewer, will send it off-list.
It crashed very early, so I think reason can be IPv6 enabled sockets or IPv6 addresses on interfaces.
Peter
On Thursday 22 of September 2005 20:59, Peter Bieringer wrote:
It crashed very early, so I think reason can be IPv6 enabled sockets or IPv6 addresses on interfaces.
I guess there's something wrong with the way I detect IPv6 resolver calls so it doesn't work with the experimental IPv6 support for Windows 2000.
I'm taking all debugging details out of the mailing list.
Best regards, Mike
-
Michal Trojnara -
Peter Bieringer