Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to email@example.com.
So after I was able to build cvsnt and install it, I tried to execute a cvs checkout command which resulted in this error: cvs init: connect to 127.0.0.1(127.0.0.1):2402 failed cvs [init aborted]: Couldn't connect to lock server After Googling around, I found this similar Windose issue that Arthur addressed in July of '06: Tony, > cvs -d C:/cvsrepos/test init > cvs init: connect to 127.0.0.1(127.0.0.1):2402 failed > cvs [init aborted]: Couldn't connect to lock server The lock server is not installed and not running, or your antivirus/firewall is preventing the connection. > The first I think is " I have the port 2401 closed" The lock service runs on port 2402 - but a "good" firewall program wont prevent a service connecting on 127.0.0.1 anyway - try opening port 2402 just to check though. Otherwise it is probably that the cvslock service has not been automatically installed and set running in the "services" windows administrative control panel. Try this if the firewall solution did not work: cd C:\Archivos de programa\CVSNT\ .\cvslock.exe -i net start cvslock Or if that fails: cd C:\Archivos de programa\CVSNT\ .\cvslock.exe -test Then in another window: cd C:\Archivos de programa\CVSNT\ .\cvs -d C:/cvsrepos/test init -a /cvsrepos-test Regards, Arthur Barrett If I manually start /usr/local/bin/cvslockd and then execute the cvs cmd, it works just fine. So the question is how do I get cvslockd to start automatically on boot up? Why didn't the installation do this? Do I have to create my own /etc/init.d/cvslock script and the appropriate /etc/rc.d/rc#.d/S##cvslockd link? - sjb - Scott Borton "Arthur Barrett" <arthur.barrett at m arch-hare.com> To Sent by: "Andreas Tscharner" cvsnt-dev-bounces <andy at vis.ethz.ch> @cvsnt.org cc No Phone Info cvsnt-dev at cvsnt.org Available Subject Re: [cvsnt-dev] build errors on Fedora 9 12/03/2008 01:01 PM Andreas, > As far as I know, gcc-4.4 is more strict, e.g. it requires > the include > file to be #included where it is needed and it does not check > if it was > included somewhere in the other included files. > A lot of changes that I had to make for Debian was because of this > behavior, which gcc-4.3 started and gcc-4.4 seems to go even further. This behaviour appears abnormal (we compile the same code with ms c++ on windows and also aCC on HPUX after all). Is it possible to change this behavious with a parameter passed to gcc? If so I'd prefer it handled that way since we rely on included files from headers within headers all over the code. > That solved the problem...put the three lines at the end of > cvsapi.h and it > now builds and installs. Thanks! But I suppose this brings up the question again of why gcc 4.4 fails to compile the code in the first place. There is no 'harm' in 'also' adding the lines suggested (however adding to cvsapi.h is preferable to adding to each trigger). Regards, Arthur _______________________________________________ cvsnt-dev mailing list cvsnt-dev at cvsnt.org http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt-dev