Hi Vinicius,
Nice approach. Like Akhil, we don't use data tables for locking, but we rely on them in an IVR to map 233 municipalities to their outage prompts, so we can update them without republishing the flow.
Christoph's point about write failures is a good one. Have you seen cases where two runs read the table at almost the same time, before either one updates the timestamp?
------------------------------
Diego
NA
------------------------------