Hi Guys,
I've HA cluster of 2 OPNsense 25.4.3.
Master node is in location "A", and backup/slave node is in location "B".
On screen You can see connection diagram from location "A" (B is the same, but different ip addresses/net on WAN interface). 0.0.0.0/0 route is advertised to opn-ints by BGP, so its always "Established". opn-ints advertise information about 10.15.3.0/24 (net "behind" opn-ints) to leafs (our core routers). There is L2 connection between opn-int01 and opn-int02 on vlan 1515. LAN and pfSync is on different physical connection (lagg1) and its always working (sorry for typos on picture).
opn.png
The problem that I have, is with lagg0_vlan1515 (10.15.3.251/24-opn-int01-master, 10.15.3.252/24-opn-int02-slave), lagg0_vlan1516 nets, on witch I have couple o VIP's (CARP).
When master node is going down for maintenance (update reboot, shutdown, etc.), VIP's and ip that is on slave node interface on opt2 (lagg0_vlan1515) or opt3 (lagg0_vlan1516) interface stop responding to ping or anything else. Just after couple of seconds I can see in opn-int02 GUI Master status of VIP's (CARP) but still nothing working for 120 sec.
After that time everything magically start working (master node (opn-int01) still down or rebooting). When master node came up, everything comes back (VIP's to master node) without any packet loss or connectivity interruption.
Do you have any clue what this 120 second "timer" could be about?
I will be grateful for any tips.
Regards
Borys