Skip to main content
Answering β€œis anything broken?” used to take three commands: px0 daemon status for whether the scheduler is alive, px0 runs list --failed for what went wrong, and px0 doctor for whether the install is sound. px0 status assembles the parts of those that matter into one screen, and makes no network or model call, so it is cheap enough to run whenever you wonder.

What it reports

--hours N

How far back a failure still counts as news.
  • Default: 24.

--json

The same information as one object: daemon state, workflow counts, next fire times, recent runs, failures, the notification policy, and the problems found.

The problems it looks for

Each problem prints the exact command that addresses it, the same discipline px0 doctor follows - a line you cannot act on is not useful output.

Exit codes

That makes px0 status usable as a cheap health check in a hook or a script, the same way px0 doctor is - but without invoking your model backend or making a network call, so it is safe to run often.

Schedules and the daemon

What the scheduler does, and how retries and failure notifications work.

Troubleshooting

Every px0 doctor check, in depth.