I have a personal OpenVPN server out on "the tubes". Last week, I was able to connect without incident and with very good performance. This week, when I manage to connect, the client seems to enter a "reboot loop".. the output as follows:
Fri Apr 13 11:20:55 2012 OPTIONS IMPORT: timers and/or timeouts modified
Fri Apr 13 11:20:55 2012 OPTIONS IMPORT: --ifconfig/up options modified
Fri Apr 13 11:20:55 2012 OPTIONS IMPORT: route options modified
Fri Apr 13 11:20:55 2012 OPTIONS IMPORT: --ip-win32 and/or --dhcp-option options modified
Fri Apr 13 11:20:55 2012 Preserving previous TUN/TAP instance: tun1
Fri Apr 13 11:20:55 2012 Initialization Sequence Completed
**Fri Apr 13 11:21:00 2012 Connection reset, restarting [0]**
Fri Apr 13 11:21:00 2012 TCP/UDP: Closing socket
Fri Apr 13 11:21:00 2012 SIGUSR1[soft,connection-reset] received, process restarting
Fri Apr 13 11:21:00 2012 Restart pause, 5 second(s)
Every few seconds, this set of messages repeats itself. Note the line I emphasized with the **. Is the local organizations's network I'm on resetting this TCP connection? The funny thing is, I can still connect to my VPN, but it's very, very slow. I think what's happening is that it gets only a few seconds at a time. Has anyone seen anything like this before, is my hypothesis true? If so, what alternatives are there? Using UDP instead of TCP, or perhaps a different port?
The problem in this case what that there was a lingering instance of the VPN client. That was the cause of the problem, however the strange fact remains that internet access was splotchy and slow (not altogether absent) while this reboot-loop was occuring.