Rumus Cloud is a container runtime with opinions already made: images are built on our builders, releases are health-checked before traffic moves, and every service gets a certificate, logs, metrics and backups without further configuration.
Each step is visible in the dashboard, and each one can fail without taking the running version down.
| Step | What happens | Typical duration |
|---|---|---|
| Detect | Buildpack inspected from repository contents, or Dockerfile used as provided | 2–4 s |
| Build | Image built on an isolated builder with a layer cache kept between releases | 35–180 s |
| Prepare | Volumes mounted, secrets injected, private network attached | 3–6 s |
| Start | Container started, port bound, readiness probe polled | 8–40 s |
| Cut over | Proxy switched to the new container, previous one kept for rollback | < 1 s |
*.rumuscloud.id or your own domainWe use the same API we teach, on real traffic. Three integrations are live today.
| Integration | Input | Output |
|---|---|---|
| Build failure analysis | Build log, last 400 lines, repository language | A one-paragraph cause and the file likely responsible, shown in the deploy view |
| Log summarisation | Filtered log window chosen by the operator | Grouped error signatures with counts and first occurrence |
| Support assistant | Question plus retrieved documentation passages | Answer with the source sections quoted, never outside the docs |