What actions are triggered by gateway going down?

[jonallport]

Apologies of this has been answered elsewhere, but I can't for the life of me find it!

What are the 'actions' that occur when a gateway monitor declares a gateway as down? And are they configurable?

TLDR: WAN DHCP does not release/renew if the upstream connection drops unless the interface is cycled or actioned from the interface status UI. How to fix?

Long version:
Netgate 1100
2x WAN connections
1x FTTP PPPoE - Tier 1
1x LTE via gateway in passthrough mode - Tier 2

The LTE is piped through the house on a transit VLAN rather than connected to the firewall as I have the firewall at ground level and want the LTE gateway as high up as possible. Gateway is in passthrough so I get the public IP on my WAN interface (and no double-NAT).

Both connections have gateway monitors pinging remote IPs. Failover FTTP>LTE works great. Fail-back works, too.

Now the problem...
FTTP PPPoE goes down: PPP client continuously retries to initiate the connection until it succeeds - good
LTE carrier goes down: interface just sits there and carries on pinging the remote IP - bad

The LTE will probably reconnect with a different IP address, so I need the interface to release / renew the DHCP lease when the gateway goes down. I've run packet capture while I kill the LTE and see no evidence of any attempt to repair/re-initialise the connection. Just carries on pinging the nominated IP in forlorn hope...

Any hints on a solution to this? Other than notifications and manual intervention

Thanks in advance

[jonallport]

It would seem the answer to my question is "/etc/rc.gateway_alarm" is run.

Nothing in there for DHCP leases from what I see. More about restarting VPN sessions and flushing states.