Hey th00ber, could you consider adding a 24h average hashrate as well as current runtime on the dashboard for each rig (I know runtime exist separately in the advanced section).
I've had the graph showing me an uninterrupted line, while the rig had rebooted sometime during that time.
Btw, the issue I've posted above, when rig is reported as offline due to some buffer overflow in the monitorig.exe still persists occasionally on several rigs. More on some than the others.