Hi,
got a question regarding failover. Would like to make sure loraserver is always available but without adding the complexity of setting up load balancers / proxies and clusters. Imagine gateways (Laird RG186) having a single destination (mqtt broker) and that we already have the means to re-route traffic automatically to a backup site when not available. If the backup site contains the same setup as the main site (mqtt broker, lora network server, lora app server), is it feasible to keep postgresql in sync using a backup mechanism and keep redis in sync by also providing the backup site with node messages so both installations always get packets from a node ? Would this not negatively influence downlinks during joins or when a command needs to be sent to a node ? What I am aiming at is an automatic failover from main to backup in case of downtime to improve availability.