tailscale / tailscale/tailscale
Can't access any port on host when tailscale is up
- Dominant language
- Go
- Stars
- 36.5k
- Forks
- 3.2k
- Avg merge
- 2d 3h
- Merged PRs (30d)
- 123
Description
### What is the issue?
I have tailscale on multiple hosts on my local network. For one host (based on proxmox, e.g. debian 12 as several are), when I enable tailscale the other hosts are unable to access any port on that host (ping, ssh, etc). When I disable tailscale access is restored. If I set -accept-routes=false, access is restored. I have two subnets set up - one for the local subnet (enabled from a VM on the host) and one on a VPS. The reason this host has tailscale enabled is so that it can access the vps subnet. The fact that it is on the local subnet (which is being advertised) is only because I don't believe that there is a way to accept just one route. I have exit nodes configured but no one is using them currently. Both the local subnet router and exit nodes are so that my phone can access them when remote. I have no idea where to start. Without access to the vps subnet, my proxmox backups will fail.
Note that since this is based on proxmox, a bridge interface (vmrb0) is configured. iptables-save looks like
```
# Generated by iptables-save v1.8.9 on Mon Nov 18 16:54:14 2024
*raw
:PREROUTING ACCEPT [52074:12918347]
:OUTPUT ACCEPT [39589:6583245]
COMMIT
# Completed on Mon Nov 18 16:54:14 2024
# Generated by iptables-save v1.8.9 on Mon Nov 18 16:54:14 2024
*nat
:PREROUTING ACCEPT [83:11651]
:INPUT ACCEPT [13:2258]
:OUTPUT ACCEPT [52:5964]
:POSTROUTING ACCEPT [52:5964]
:ts-postrouting - [0:0]
-A POSTROUTING -j ts-postrouting
-A ts-postrouting -m mark --mark 0x40000/0xff0000 -j MASQUERADE
COMMIT
# Completed on Mon Nov 18 16:54:14 2024
# Generated by iptables-save v1.8.9 on Mon Nov 18 16:54:14 2024
*filter
:INPUT ACCEPT [102:49953]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [547:110867]
:ts-forward - [0:0]
:ts-input - [0:0]
-A INPUT -j ts-input
-A FORWARD -j ts-forward
-A ts-forward -i tailscale0 -j MARK --set-xmark 0x40000/0xff0000
-A ts-forward -m mark --mark 0x40000/0xff0000 -j ACCEPT
-A ts-forward -s 100.64.0.0/10 -o tailscale0 -j DROP
-A ts-forward -o tailscale0 -j ACCEPT
-A ts-input -s 100.69.198.12/32 -i lo -j ACCEPT
-A ts-input -s 100.115.92.0/23 ! -i tailscale0 -j RETURN
-A ts-input -s 100.64.0.0/10 ! -i tailscale0 -j DROP
-A ts-input -i tailscale0 -j ACCEPT
-A ts-input -p udp -m udp --dport 44647 -j ACCEPT
COMMIT
# Completed on Mon Nov 18 16:54:14 2024
```
Any ideas?
### Steps to reproduce
Access to the host with tailscale enabled was working until I rebooted the host. Now access requires either tailscale to be down or for --accept-routes=false.
As stated before, note that this is running proxmox with the proxmox kernel and a bridge interface configured. I'm assuming that either or both are significant but don't know why.
### Are there any recent changes that introduced the issue?
Not that I am aware of
### OS
Linux
### OS version
Debian 12, proxmox 8.2.8
### Tailscale version
1.76.6
### Other software
_No response_
### Bug report
BUG-da57ac32d3f694fa43d53a247618a6c8dd9a28f70085742e8167ba84fb84f04d-20241118213040Z-8b6e889586be8270
Contributor guide
Assessment
This issue has not been assessed yet.