-
Notifications
You must be signed in to change notification settings - Fork 759
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ikev2 ipsec roadwarrior with virtual subnet configuration issue #2334
Comments
|
Hi, 3c3628c should fix the issue, with ikev1 it worked with ikev2 apparently not. Can you try and confirm? |
|
It does, thank you. The generated config file now looks like so: did you mean to leave that rightsubnet statement empty? I am not sure if that would be a problem or not. |
|
oops, missed a spot apparently. eaf1927 solves the empty rightsubnet when tunnel isolation is enabled. |
|
works as expected, thanks! Without tunnel isolation (which I enabled during debugging), this is what the generated file looks like: |
|
thanks for reporting back! |
… see opnsense/core#2334 (cherry picked from commit 3c3628c) (cherry picked from commit eaf1927)
Hi there,
I am using opnsense as an IPSec VPN access point i.e. opnsense is configured as a responder only, with a roadwarrior/mobile client ikev2 configuration, as shown in https://docs.opnsense.org/manual/how-tos/ipsec-road.html . After authentication, each client gets an ip address from the virtual pool.
This set-up works fine with a single client. As soon as a second session is brought up (by a different user), traffic on the first client stops. Both clients authenticate just fine, security associations are set up (and stay up), but packets are only passed to the last client to have connected.
Logging in through SSH and poking at configuration files, I tracked down the issue to be an ipsec.conf problem:
From https://wiki.strongswan.org/projects/strongswan/wiki/VirtualIP :
Removing "rightsubnet = 192.168.0.0/24" and restarting strongswan solved the issue: multiple roadwarriors can now simultaneously connect and their traffic is correctly routed.
Could that be an ipsec.conf generator problem, or did I miss a setting somewhere?
This is on the latest release (18.1.5, i386/openssl), on an alix2d3 board (not the fastest but it gets the job done).
I am not sure unfortunately. I started using opnsense around last december and I believe the bug has always been present (VPN flakiness has always been reported but I haven't had time to dig in until now)
Like i said above, I believe the bug was already present when I started using opnsense.
That would be the IPSec configuration pages. VPN > IPSec > Mobile client/Tunnel configuration.
to restart strongswan.
7) disconnect and reconnect both clients, notice that they both get inbound traffic.
Optionally before step 6, type
to inspect ipsec policies and see that with the rightsubnet statement, the entire virtual pool is routed to every client, thus explaining the routing issue i've been seeing.
The text was updated successfully, but these errors were encountered: