IPsec EAP-TLS can't reach remote network
-
I have found in my IPsec logs that i have a DPD error, but I can't see where it's coming from or how to solve this
-
Tbh I don't see any DPD error there, they're just standard log messages. I don't see anything that looks related to the issue you're having.
I'm not sure I can suggest anything useful here, but if I understand your setup looks something like:
client <-> IPSEC <->pfSense <-> 172.16.1.0/22 LAN
And that 172.16.1.0/22 LAN is directly connected to your pfSense box?
What phase 2 network did you assign?
How big is the 10.10.0.0 subnet? /24? Is it definitely not used or overlapping with anything else the pfSense box can reach? I understand overlapping networks cause all sorts of problems in IPSEC.
Also, I notice in your screenshot, your local Ethernet subnet mask is 255.255.252.0 which is a /22, but you said it was a /24, which should be 255.255.255.0. I'm not sure if that's relevant.
FWIW, I don't get a gateway on my IPSEC interface either, 0.0.0.0 seems normal.
-
You may find setting some of the IPSEC services to the diag log level as described here:
https://docs.netgate.com/pfsense/en/latest/vpn/ipsec/ipsec-troubleshooting.html#ipsec-debugging
From what I understand, your tunnel is coming up, but you can't reach anything connected to the remote pfSense networks once connected. It's hard to say what that could be, since it worked briefly and then stopped after a reboot. Are you pinging by IP or hostname?
-
Hi,
Alright I thought it a DPD error my bad then.
Yes my setup is exactly this one. Yes all my LANs are directly connected to my pfSense.
The network assigned is my Phase 2 is my 172.16.1.0/22 LAN (called LANSERVEURS as you can see)
It is indeed a /24 network but also my only network using 10.10.x.x so no risk of overlapping anything other subnet.
Yes my bad, both LANs are actually in /22 so having a .252, I've wrote to fast this information sorry. But everything is correct in the configuration.
Yes it exactly do that. And sometimes with no reason my tunnel is fully functionnal for a few minutes with traffic on IPsec interface logs(ping, acces https web console on distant machines, etc)
-
Are you attempting to reach your hosts by IP or hostname?
Also, I notice that 172.16.1.0/22 would actually include anything between 172.16.0.0 and 172.16.3.255. 172.16.0.0/24 wouldn't be a separate network. Do you have anything else using 172.16.0.0/24?
Can you tcpdump on pfsense while pinging from your VPN client to see where traffic gets to? Example:
tcpdump -i enc0 src 10.10.0.1
tcpdump -i <name of 172.16.1.0/22 interface> src 10.10.0.1
-
By IP only for now
Yes it does include that range, though I don't have anything using 172.16.0.0/24. My first network starts at 172.16.1.0/22.
Nothing is shown on the tcpdump. As I could see on my firewall traffic, there is nothing after the tunnel connection to my interface (port 500). No sign of ICMP traffic
-
So can you see traffic actually hitting your IPSEC enc0 interface on the pfSense end at all or nothing?
Can you ping an IP on the pfSense box?
The Windows IPSEC client ignores static routes sent by a VPN gateway. Did you set the VPN connection on your Windows client to 'use default gateway on remote network'?:
- Open Network and Internet Settings on the client PC
- Click on the VPN tab on the left
- Click Change adapter options on the right hand side of the VPN tab
- Find your VPN connection, right click it and click Properties
- On the Networking tab, select IPv4 and click properties, then Advanced
- Tick Use default gateway on remote network
- Click OK
-
There are 2 scenarios :
-
First one, is that I see only the traffic of the connection from the client to my tunnel interface, the tunnel is up but no traffic is seen after from the tunnel IP pool (as it was not up).
-
Second one, the tunnel goes up and everything works fine BUT it last only for a few minutes (sometimes randomly goes back up again and same thing)
For your questions :
-
Yes I do see the IPsec traffic connection :
-
You mean ping the client or remote IP from the pfSense ? In both cases I the pfSense can reach everything.
-
Yes I have noticed this parameter before and I already checked it. I first thought it was coming from the VPN gateway too.
But I just noticed that the default gateway of the pfSense is my NAT gateway (my all lab is behind a NAT), so for my client using this gateway won't be any use. I don't know if i'm clear enough or a diagram of my architecture is needed.
-
-
Anyone ? :/
-
I'm out of ideas to be honest. If you're running the latest pfSense and the VPN is coming up but sometimes no traffic is passing, sometimes it is then I'd be inclined to think it's not pfSense but an external factor, so what's different on the client and on pfSense between when it works and when it doesn't?
Is some other VPN up? Are you using wifi instead of wired? Can you access the Internet when the VPN is up? Is something else different? Have you look at the Windows Event Viewer and the pfSense logs in both scenarios?
I meant from the client, can you ping interfaces on the destination pfsense box, other than the IPSEC interface? For example can you ping the pfsense interface that is connected to the LAN that you're trying to reach? What about other interfaces on the pfsense box?
You're right, I don't really understand what you meant in the last sentence about the NAT gateway. A diagram may help me understand, but as I said, I'm not sure how much more I can help diagnose the problem.
-
@ads76 I also think it's an external factor, like the Windows client but nothing in the Event viewer logs appart the connection setting up.
I don't have another VPN no. It's a virtual lab built on an ESXi so I'd say it's wired-like. No I can't while it's up. pfSense logs only show this (as shown before) :
When the tunnel is up I cannot access or ping anything outside the LAN where the client is and the WAN interface (192.168.101.40 - tunnel).
We don't need to bother about the NAT gateway it's irrelevant sorry it's juste the default gateway of my pfsense to go outside of my lab.