Looks like I've found the culprit for your issue. I discovered it when I created a new Wireguard tunnel for a router and the client wasn't making any connection attempt.Im having Wireguard issues as well but I havent needed to use the "Responder" settings on either side. This is in a S2S configuration where my home RB5009 is on 7.19.1 and work production CCR1009 is on v7.18.2Resolved Wireguard issue. It was related to the /interface/wireguard/peers "responder" setting. It seems MT changed which side it needed to be applied to, as I had to swap it from the client peer definitions to the server peer definitions for the VPNs to work again.
EDIT: Im getting the below log messages on both routers.EDIT2: Just reverted back to v7.18.2 on RB5009 and Wireguard S2S connectivity is restored.Code:wireguard-DC-S2S: [DC-S2S] Handshake for peer did not complete after 20 attempts, giving up
Turns out that (and this is probably a bug introduced in 7.19.x, if it's not intentional), you must now add the "persistent-keepalive" option to at least your client-side peer; otherwise, no connection attempt is made.
I'm going to go ahead an file a bug report, as this is a regression from pre-7.19.x behavior and the "persistent-keepalive" option is not being treated as mandatory.
Statistics: Posted by rajo — Sat May 31, 2025 5:05 am