Re: [Licq-devel] another trace

2005-12-02 Thread Philipp Kolmann
On Fri, Dec 02, 2005 at 07:32:46AM +0900, Jon Keating wrote: > On Wed, Nov 30, 2005 at 10:39:44PM +0100, Philipp Kolmann wrote: > > just tested it with someone with LICQ 1.3.0/SSL > > > > There the SSL Tunnel works. So it seems that the receiving end has the > > problems. > > > > just a thought..

Re: [Licq-devel] another trace

2005-12-02 Thread Daniel Glöckner
> Yes, please do. So far everyone is using 0.9.7 and having this error, so > perhaps it is an OpenSSL issue. I'm using Licq Daily 2004-06-26 (yes, no typo). It _works_ with OpenSSL 0.9.7f <-> 0.9.7f 0.9.7g <-> 0.9.7g 0.9.7f <-> 0.9.7i 0.9.7f <-> 0.9.8a 0.9.8a <-> 0.9.8a between different users on

[Licq-devel] Re: MSN "Winks" support?

2005-12-02 Thread D�niel Fraga
In article <[EMAIL PROTECTED]>, Jon Keating <[EMAIL PROTECTED]> writes: > You receive a message from MSN or the user? Yes. It says something like that everytime a user tries to send me a "wink": :(l) Tania sent you a wink which requires the latest version of MSN Messenger. To dow

Re: [Licq-devel] another trace

2005-12-02 Thread Ian Collier
On Fri, Dec 02, 2005 at 04:41:14PM +0900, Jon Keating wrote: > Yes, please do. So far everyone is using 0.9.7 and having this error, so > perhaps > it is an OpenSSL issue. > > I've asked others to test different versions, but no one has reported back to > me about it. I am running Licq 1.3.0 com

[Licq-devel] OpenSSL Connections

2005-12-02 Thread Jon Keating
After examining the code, it looks like the problem was introduced around May 20th. The current CVS has the proper fix in there in reality it was just a one line fix. Enjoy your SSL again... Jon pgpEwww7Ftc7s.pgp Description: PGP signature

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Ladislav Michnovič
Can you post this patch, please, if it is only one line. I don't want to download a whole CVS version and look for it. Maybe I'm not the only one. Thank you. Regards Ladislav. 2005/12/2, Jon Keating <[EMAIL PROTECTED]>: > After examining the code, it looks like the problem was introduced aroun

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Jon Keating
On Fri, Dec 02, 2005 at 04:20:27PM +0100, Ladislav Michnovi? wrote: > Can you post this patch, please, if it is only one line. I don't want > to download a whole CVS version and look for it. Maybe I'm not the > only one. It's on Licq-CVS... Jon Index: icqpacket.cpp

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Ladislav Michnovič
2005/12/2, Jon Keating <[EMAIL PROTECTED]>: > It's on Licq-CVS... Thank you. > --- icqpacket.cpp 7 Oct 2005 16:34:01 - 1.124 > +++ icqpacket.cpp 2 Dec 2005 15:00:02 - 1.125 > @@ -4827,7 +4827,8 @@ > m_nSize -= strlen(m_szMessage); ^^^

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Jon Keating
On Fri, Dec 02, 2005 at 04:55:30PM +0100, Ladislav Michnovi? wrote: > m_nSize += strlen(m_szMessage); is replaced by > m_nMsgLen = strlen(m_szMessage) + 1; etc. > why not the line above ? > It is only my first look, don't bother to explain why, if it is O.K. > Just say it is O.K. Both are fine, I

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Philipp Kolmann
On Sat, Dec 03, 2005 at 12:01:18AM +0900, Jon Keating wrote: > After examining the code, it looks like the problem was introduced around > May 20th. The current CVS has the proper fix in there in reality it was > just a one line fix. I have applied the patch, but still can't receive a connect

Re: [Licq-devel] OpenSSL Connections

2005-12-02 Thread Jon Keating
On Fri, Dec 02, 2005 at 06:54:51PM +0100, Philipp Kolmann wrote: > I have applied the patch, but still can't receive a connection from a 1.3.0 > client: Try the CVS daily. It's being built now... 20051203 Jon --- This SF.net email is sponsored

[Licq-devel] SSL problem seems fixed

2005-12-02 Thread Pollywog
The SSL problem does seem to be fixed for me. I tried a SSL connection with one person and it worked for me and for him. thanks, Jon. 8) --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop!