# Recovery Scenarios

> Use resume (CLI or PM2 restart) after resolving the issue. Check audit and verbose logs to see the exact failure point.

Source: https://bifrost5.com/foundry/docs/recovery-wiki/

# Recovery Scenarios

Information

Get back to a working setup with one command from encrypted cloud backup, with an engineer to help, in Production-Grade.

[See editions](/foundry/docs/editions)

| Scenario | What Happens |
| --- | --- |
| **Provider CLI crash** | Reload rules and state; reissue the last task. |
| **Supervisor restart** | Load last persisted state from DragonflyDB; resume from next task. |
| **Task blocked** (failed after max retries) | Task marked blocked; operator must clarify or adjust; then resume. |
| **Conflicting state** | Halt; operator resolves (e.g. clear or fix state); then resume. |
| **Persistence failure** | Halt immediately; fix DragonflyDB/Redis; then restart. |

Use `resume` (CLI or PM2 restart) after resolving the issue. Check audit and verbose logs to see the exact failure point.