Hello,
I am running into a odd problem I am not sure how to solve when isolating subnets using zone policies on my ER-X. I have created 3 vlans, a main network, a management network and a guest network. The management network is VLAN 100 on the 192.168.100.0/24 subnet and the guest network is on VLAN 20 and the 192.168.20.0/24 subnet.
I have eth 1-4 as part of switch 0, and eth0 connected to the internet. swith0.20 has an address of 192.168.20.1 and swith0.100 has an address of 192.168.100.1.
The issue I am running into is that while I am connected to the guest network I am able to 192.168.20.1, which makes sense. But I am also able to ping 192.168.100.1. This makes sense to me because it is also technically local, but this is not the result I am looking for. Since 192.168.100.1 is on a different subnet, I would not want to be able to ping it.
Below are my configurations for my local zone and the offending firewall rules.
zone local { default-action drop from guest { firewall { name guest-local } } from house { firewall { name lan-local } } from management { firewall { name management-local } } local-zone }
default-action drop enable-default-log rule 1 { action accept state { established enable related enable } } rule 2 { action drop log enable state { invalid enable } } rule 100 { action accept protocol icmp } rule 600 { action accept description "Allow DNS" destination { port 53 } protocol tcp_udp } rule 700 { action accept description "Allow DHCP" destination { port 67,68 } protocol udp }
I attempted to add in another rule to drop all connections to the 192.168.0.0/16 subnet, but this caused me to be unable to ping 192.168.20.1. I thought of creating a network group that contained all of the 192.168.0.0/16 subnet except the 192.168.20.1/24, but this seemed messy to implement and I figure that has to be a cleaner method to do this.
This issue exists for the main network as well, but I figure once I find a solution between these two VLANS I can simply transfer it over for the remaining rule sets.