Detects crash loops in container services by counting recent exit events.
// Source: https://github.com/InsForge/InsForge | License: Apache-2.0 | Copyright You must retain, in the Source form of any Derivative Works // Provided via justclaw.it.com; keep this notice with the code. // Derive a simple crash-loop signal from container lifecycle events. // // A healthy machine reports a single `start: started` after launch and stays // quiet. A boot-looping container (e.g. Node throws on import → Fly restarts // → throws → restarts) emits a steady cadence of `launch: pending` → // `start: started` → `exit: stopped` triples roughly every 10s. // // The two drivers name the same moment differently: Fly reports `exit: stopped`, // the Docker driver passes the daemon's own action through as `[container] die`. // Matching only Fly's wording meant a restart-looping Docker ... (truncated -- full source via MCP)
See the full source, get the GitHub permalink, and search for more like it.
Get a free API key