Understand the results
Inspect runs
A job defines recurring work. A run records one occurrence. Open Runs to browse executions and reports, or use a job’s runs tab to narrow the history.
Find the run
Filter runs by job kind, status, machine, job, and date. Command jobs show the machine that ran them; monitored jobs report without a machine.
The overview brings together active executions, upcoming jobs, and items that need attention. Open a run to inspect what happened.
Understand the status
| Status | Meaning |
|---|---|
| Queued | Work is waiting to be dispatched. |
| Dispatched | Work has been assigned to an agent. |
| Running | Execution has started, or a monitored job sent a start signal. |
| Success | The command or report indicated success. |
| Failed | The command exited unsuccessfully, or the work reported failure. |
| Timed out | The execution reached its allowed time limit. |
| Cancelled | The run was cancelled. |
| Lost | The server could no longer confirm the execution’s state. |
| Missed | Scheduled work did not run or report inside the expected window. |
Read the explanation on the run page alongside its status. A missed occurrence because a machine was offline is different from a missing ping; a lost run does not establish whether the underlying command finished.
Read the output
Open the run’s output tab. You can search, copy, or download its logs. Command runs can stream output while executing and retain it afterwards. Monitored runs can include output submitted with a report.
Output may be capped or pruned according to configured limits and retention. Avoid writing credentials or other secrets to it.
Inspect the timeline
The timeline tab lists recorded events in the run’s lifecycle. Use it with the start and finish times, duration, exit code, and status explanation when investigating a problem.
Runs preserve the job definition used for that occurrence, so edits to a job do not change the historical command.
Run again or cancel
For a completed command run, Run again creates another attempt. It does not overwrite the previous one. Correct the cause of a failure before retrying work with side effects.
For active command work, cancellation requests the execution to stop. Monitored work runs outside Oh My Job, so the server cannot cancel or rerun the external process.
Decide what happens when a machine is offline
Command jobs can skip missed work or allow it to run late within a grace period. Multiple missed occurrences are coalesced rather than replayed as an unlimited backlog. Set the policy according to the command: running a delayed backup may be useful; replaying stale work may not be.
See the operations reference for detailed scheduling behavior and maintenance controls.