After recent incidents with the datacenter SetCronJob hosted in, I decided to move all the main servers to another data center as well as work on multi-datacenter clustering.
The servers transfer and multi-datacenter setup were completed without any service downtime. Now SetCronJob fully achieved high availability that can withstand whole datacenter outages.
The clustering was done with great help from Clustermin, my developing service for server cluster management and automation. Setting up once and it'll run setup script, update firewall rules and configurations files in all servers and all clusters, whenever a new server is added or old server removed.
The current setup will greatly improve our service reliability, and hopefully, I don't need to inform you about any downtime again ;-)