Customer spoke site (SiteA) has two internet wan links as shown in below diagram. WAN1 has additional hops before breaking out to internet.
LIBO traffic (local brekaout) load balance on both WAN links. Whenever there is an link issue between the ISP1 provider's PE and the router (ISP1_GW) at SITEA-NODE1 location, LIBO traffic is still sent towards ISP1_CPE router . As per CSO design RPM probe and ip-monitoring policy monitors connectivity with the immediate next hop . Due to this traffic will get blackholed at ISP1_GW router when there is a link failure between ISP1_core router and ISP1_GW router.
This article provides the sample configuration for rpm probe and ip-monitoring policy for this scenario.
Here is the sample configuration that can be applied in the SRX CLI for the topology shown in description. With the below configuraiton whenever there is link failure between ISP1_PE router and ISP1_GW router default route metric pointing to ISP2_Core router (WAN_0 GW) will be changed to attract all LIBO traffic.