App enters crash-loop after exceeding free-tier RAM limit

When a Discloud-hosted application process exits unexpectedly, the platform attempts automatic restart; repeated rapid failures can leave the app in a crash-loop state visible through the dashboard status. This goal examines how the platform handles sustained resource exceedance within that restart cycle.

Key constraints include the free-tier RAM and CPU caps, which commonly trigger process termination, and the asynchronous nature of status updates, meaning dashboard reflections are not instantaneous. The uncertainty involves whether restart attempts are bounded or continue indefinitely, and how the platform differentiates transient spikes from sustained exceedance.

What mechanism decides when restart attempts are ceased? Does the dashboard present a distinct status or summary when the limit is reached, or is diagnosis reliant solely on log retrieval? How does the platform distinguish between a brief memory spike and a sustained exceedance in its restart policy?