Tunnel endpoint IP is excluded from tunnel?
TL;DR:
One IP address that should be included in a tunnel is instead routed to local default route.
All others are tunneled as expected. This behavior started suddenly 3 days ago after working as expected for almost a month.
MacOS warp client
-------
I have a tunnel set up with a pair of hosts running cloudflared for redundency. I am running warp ZT on MacoOS for the client side. It's all working as expected with one exception:
3 days ago direct ssh to one of the tunnel (tunnel2) hosts stopped connecting. The other (tunnel1) is still accessible and I can reach and examine tunnel2 by logging into tunnel1 and then across to tunnel2.
What I see if I traceroute to it is that the IP is excluded from the tunnel despite no exclude rule for the host or IP address:
While the other tunnel host gets on traceroute response (because the tunnel doesn't pass ICMP)
As far as I can tell the tunnel daemons are working fine and the tunnel service and the network profile are fine and al the users are happy.
When I try to ping tunnel2 I get a different response than tunnel1. (tunnel1 gets a response from
2 Replies
SIGH. I figured it out I think. The excluded destination address is the IP I configured for a known network beacon host. If that was accessible from my home network it would look like I was at that destination so it's blocked.
I need to create another IP on the destination network for the beacon so they don't conflict.
Do you have a starting point for me on configuring the IP address a tunnel publicly uses? Sounds like you've figured that out, I'm still trying to understand how that works 🙂