|
|
On Monday 09 August 2004 19:57, Erik Anderson wrote: > Sorry for another logfile analysis request, I would just appreciate some > additional clarification/nuancing over this logfile sequence: > > Aug 9 17:33:58 knight openvpn[18671]: cpm-t30/1.2.3.4:5000 [cpm-t30] > Inactivity timeout (--ping-restart), restarting > Aug 9 17:33:58 knight openvpn[18671]: cpm-t30/1.2.3.4:5000 MULTI: > multi_close_instance called > Aug 9 17:34:09 knight openvpn[18671]: TLS Error: Unknown opcode (6) > received from 1.2.3.4:5000 "Unknown opcode" occurs when the server has closed the client instance object (so it has no memory of the client connection) but is still getting stuff from the client, which it doesn't know what to do with. In the next beta, I'm going to rename that message to something more informative like the previous sentence. > Aug 9 17:34:24 knight openvpn[18671]: TLS Error: Unknown opcode (6) > received from 1.2.3.4:5000 > Aug 9 17:34:40 knight openvpn[18671]: TLS Error: Unknown opcode (6) > received from 1.2.3.4:5000 > Aug 9 17:34:44 knight openvpn[18671]: MULTI: multi_create_instance called > > Thank you for your patience with this. I have discovered that only one > OpenVPN client session can be established at a time from the network > identified as IP address 1.2.3.4 above (the machine immediately stops > responding to pings as soon as the second machine successfully logs in). I > realize that my issues may not be technially be caused by OpenVPN, but the > UDP protocol involved is definately causing some strange voodoo to occur. If you are using the same cert for both clients, make sure you have --duplicate-cn enabled. Otherwise, OpenVPN will kill the client-instance of a connection if a new connection using the same certificate tries to connect (the reason for this behavior is discussed in the archives). James ____________________________________________ Openvpn-users mailing list Openvpn-users@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/openvpn-users |