Route showing up I did not ask for.
-
Is their anything that would add / update routing tables without explicitly being put into the firewall?
The back story:
Location #1 - 192.168.0.254/24 - servers, port 2 of four port NIC
Location #1 - 10.0.2.1/24 - lan (users), port 1 of four port NIC
Location #1 - WAN - Port 0 , four port NIC
OpenVPN to Location # 2Location #2 - 10.0.1.1/24 - LAN, port #1 on four port NIC
Location # 2 - WAN , port #0 on four port NICThe above has worked great for many months (14+). I can't say enough about how consistent this has been.
The change that was made:
I moved the Location #1 servers to Location #2.
Not much had to change.
Disable Location #1 Port 2 interface
update routes on OPENVPN gateways.
Use Location#2 Port 2 NIC interface and setup same subnet.I did learn about the route flush command that night! :)
Still with me?
This worked fine for almost two months.
I had someone call and complain about no connection to the servers. The VPN was saturated to a printer! I bounced the VPN and fixed? For a few minutes.Then it went sideways. 192.168.0.7 was using ~256kb up and down on the vpn… The attached route table showed the problem... Somehow this server (windows) was added to the route table (loc#2) to go over the VPN when it was local.
The Location #1 route table showed the same. So it was ping pong back and forth till it TTL'ed....Can something else update the route tables?
What I ended up doing was using the route flush command and then adding back (route add) each route by hand on both boxes.
This morning I rebooted Location # 2 and the route table showed
192.168.0.0/24 | OpenVPN ip gateway <--- good
192.168.0.7 | OpenVPN ip gateway <---- ?? why!I'm looking for direction and or understanding as to what could possibly cause this.
Route flushing and restarting the vpn and gateways would bring back the problems. Manually deleting / adding the route tables was the only way to make it work. I fear a reboot will break it again.
Thanks!.png_thumb)
.png)