Essentially I would like to configure my firewall to Only allow IPSEC/L2TP clients into the LAN, and to also specify what they can access. For arguments sake, we’ll say an intranet web server. Is this possible? What needs to be done?
My understanding which is also applied in the above thread, is that IPSEC packets appear to arrive unencrypted on the WAN interface, so you can’t distinguish between packets from the VPN and the public internet? Is this correct?
I don’t quite follow I’m afraid, but thanks for your input! I know you can create a permanent interface on the L2TP server for a user even though the connection may not be active, is that what you refer to as binding? What if someone were to send traffic over IPSEC that wasn’t encapsulated in L2TP, wouldn’t using L2TP interfaces leave that open to attack?
I’ve not seen the bridge parts before, and having a quick flick through the wiki hasn’t made me any wiser.
Sorry if I’m being dumb, I’m trying to brush up on my overview of VPNs etc. I’ve been using routers for a few years now, but this is my first real play with L2TP / IPSEC.
Look into the bridge, incoming-filter, and outgoing-filter options. See: http://wiki.mikrotik.com/wiki/Manual:PPP_AAA#User_Profiles … I can post more later and see if I can come up with an example. But thats where you want to start.