Tailscale route shares subnet with LAN

Hi,

I have a tailscale client exposing a remote 192.168.1.0/24 subnet. Unfortunately, the pfSense I wish to connect it to is also set on that same subnet.

I therefore created a local VLAN 192.168.18.0/24 but I can’t seem to get the remote 192.168.1.0/24 into the routing table. Presumably because this is happening at the kernel level?

I really don’t want to change either subnet to something different if it can be avoided, so is there a way to route the tailscale traffic JUST through that VLAN?

Thanks!

Should also mention that the remote client is a Windows PC with the tailscale client on.

If there’s another solution to solve this problem, I’m open to it. For example, can wireguard achieve this with a windows client? Given it exists in kernel mode on pfsense

hi,

if u were running pfsense with tailscale addon, cud prob do it on this outbound nat page as in my post here … Tailscale pfsense outbound setup issue?

no replies there either to my problem :slight_smile:

i dont think this forum is the place for tailscale enquiries, going to repost mine at https://forum.tailscale.com/ and/or the pfsense forum

personally … i’d change one of the subnets no matter how inconvenient, my philosophy… KISS (keep it simple stupid :slight_smile: )

Wait, the Tailscale site been in read only mode since July '23. Just when I made quick bookmark about a forum for them a little bit ago, obviously didn’t read it though.

So I guess it is Reddit then for Tailscale, damnit.

There really is a lot of info on Reddit, a lot, but I think forums are really the way to go for searching and looking through data. Especially for tech, medical type stuff where you want to copy and paste and retain it. Actually, more I went into different topics, the more of a cluster it really is.
I guess I need to look into some kind of system to start to grab this tech info better so I can come back to it on my computer(s). I guess this is where AI could come into play. WTH did it get so big again? I need to look this up real quick.

I don’t think there is any solution to the issue other than either having different subnets for each end or loading Tailscale on each device so the subnet they are attached to does not matter.

of forgot they had, thought it was a real error on their side to do that :frowning: