Hi there,
I’m noticing a high number of connection drops when 2 RB’s are VPNing between each other. But oddly enough as soon as its drops, it will reinstate the connect and work as normal. After a few minutes the whole process happens again.
What is the best method to find out the cause of these drops?
VPN sever - x86 (cloud hosted) - 6.15
VPN client - RB450G / RB1100AHx2 - 6.7
Ok it seem looking at the logs its disconnecting very 65 seconds exactly.
PPTP has a keepalive feature where one end sends an echo request and the other must reply. 60 seconds sounds like a likely timeout period. You might check configuration of the PPTP at both ends and see if there’s any setting for that. Failing that, look for any other PPTP settings to tweak, such as header compression, which can be tricky (although I’d expect ROS’s implementation to be able to talk to itself).
ROS Packet Sniffer doc mentions gre, which is used to carry PPTP traffic, so checking that might provide a clue. Control traffic is carried on TCP port 1723 in standard PPTP protocol (I don’t know if that applies here). I usually use Wireshark to look at network traffic (easier said than done).
I’m brand new with RouterOS, but I thought I’d take a stab at this since I have some background with PPTP and this wasn’t answered. Good luck!
On second thought, keepalive may be working correctly (hanging up after 60 seconds) and your PPTP payloads aren’t getting through. Can you ping over PPTP? Here’s a thread that may help: http://forum.mikrotik.com/t/pptp-problem-with-tunnel/79090/1