You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
After upgrading to OPNsense 25.1-amd64, the NTP service fails to start when a WireGuard instance is configured on the same physical interface that NTP is set to listen on.
Unfortunately, I am unsure of the exact previous version, but it was from the 24.x branch. I attempted to use opnsense-revert to determine the last known working version but was unable to retrieve this information.
To Reproduce
Steps to reproduce the behaviour:
Create a wireguard instance on an interface (re0_vlan225 - Lab)
In the NTP Service setting assign the interface (Lab)
Click apply
Start the service
The service does not start. The log contains:
2025-02-18T10:00:00 Error ntpd unable to create socket on wg0 (15) for 172.29.0.1:123
2025-02-18T10:00:00 Error ntpd bind(35) AF_INET 172.29.0.1:123 flags 0x11 failed: Address already in use
2025-02-18T10:00:00 Error ntpd daemon child died with signal 11
Expected behavior
The NTP server should start
Describe alternatives you considered
I tried unassigning the wg0 interface to see if that would work but get the same error.
Screenshots
NA
Relevant log files
NA
Additional context
Add any other context about the problem here.
NA
Software version used and hardware type if relevant, e.g.:
OPNsense 25.1 (amd64).
The text was updated successfully, but these errors were encountered:
This applies to any kind of service that you try to bind to specific interfaces, especially vpn ones. Use any interface instead and control access via firewall rules.
Thinking about it I got my logic wrong because I don't think wireguard binds to any physical interface it's purely virtual. I am not sure why the Time Services want to open a port on it since it is not available as an interface to bind to in the config ui.
@Monviech I don't select the wireguard interface to have NTP services bound to in the config UI. The configuration attempts to start an NTP server on it by itself for some reason.
@fichtner It works for a while before the NTP server fails again. Still investigating it. Will check the config file.
Important notices
Before you add a new report, we ask you kindly to acknowledge the following:
Describe the bug
After upgrading to OPNsense 25.1-amd64, the NTP service fails to start when a WireGuard instance is configured on the same physical interface that NTP is set to listen on.
Unfortunately, I am unsure of the exact previous version, but it was from the 24.x branch. I attempted to use opnsense-revert to determine the last known working version but was unable to retrieve this information.
To Reproduce
Steps to reproduce the behaviour:
2025-02-18T10:00:00 Error ntpd unable to create socket on wg0 (15) for 172.29.0.1:123
2025-02-18T10:00:00 Error ntpd bind(35) AF_INET 172.29.0.1:123 flags 0x11 failed: Address already in use
2025-02-18T10:00:00 Error ntpd daemon child died with signal 11
Expected behavior
The NTP server should start
Describe alternatives you considered
I tried unassigning the wg0 interface to see if that would work but get the same error.
Screenshots
NA
Relevant log files
NA
Additional context
Add any other context about the problem here.
NA
Software version used and hardware type if relevant, e.g.:
OPNsense 25.1 (amd64).
The text was updated successfully, but these errors were encountered: