I have two wg(4) interfaces: one that is a site-to-site tunnel (i.e., exactly one wgpeer where both sides have wgendpoint configured) in rdomain(4) 1, and another that is used as the "server" for roaming VPN clients in rdomain(4) 0. Last week I ran ifconfig wg1 destroy, replaced the wgkey and wgpsk for one of the three wgpeers in the second interface, and ran sh /etc/netstart wg1. Once I did this, the server seemingly froze: it stopped responding to ICMPv6, ICMP, and UDP traffic sent to a vlan(4) interface in rdomain(4) 0 whose parent is an ixl(4) interface. After about a minute, the server started to respond to traffic; nonetheless I rebooted the server as a precautionary measure.
I almost never change the wg(4) config, so I cannot say if this is a new issue or an issue at all-perhaps it was just a coincidence-and if an issue, not sure if this is related to what is happening to you. I thought I'd provide a possible data point though. The server is running OpenBSD 7.4-stable.
