Skip to main content
stackdome status answers two different questions: which release is serving, and whether that release is healthy now. Use structured output for automation and the table view for resource-level investigation.

​
Check the current stack

The table shows the serving release state followed by every resource, its runtime state, declared ports, and public URL. When the API reports a failure, the CLI also prints its captured details and the three most recent resource conditions. To inspect a stack other than the current context, pass its name:
The current CLI declares an optional [resource] argument, but status output remains stack-wide. Use stackdome logs <resource> when you need to focus on one resource.

​
Prove convergence and health

For an agent or CI job, request JSON:
The result is the Stack object. Its two embedded release summaries have different meanings: A stack is serving a healthy release when all of these are true:
  • converged_release is present.
  • converged_release.state is Released.
  • converged_release.health is ok.
To verify a deployment you just created, retain release.id from stackdome deploy --wait -o json as the deployed release ID and also require converged_release.id to equal it. This prevents an older healthy release from satisfying the check. To prove that this deployed release is also the newest attempt, require:
  • latest_release.id equals the deployed release ID.
  • latest_release.state is Released.
Do not require latest_release.health to be ok. That field is used for a newest attempt that is still progressing or has failed; once the attempt has settled successfully, runtime health belongs to converged_release.health. The possible runtime health values are ok, progressing, degraded, unavailable, and failed. ok is the only value that proves the serving release is healthy.

​
Watch a rollout

The CLI refreshes every three seconds until interrupted. In table mode it redraws an interactive terminal. In structured mode it emits one complete Stack object per refresh:
This is a stream of JSON values, not one JSON array. When watching a deployment, parse each value independently and stop only after converged_release.id equals the retained deployed release ID, its state is Released, and its health is ok. If the watcher also claims this remains the newest attempt, require latest_release.id to equal the deployed release ID and latest_release.state to be Released.
Interrupting stackdome status --watch currently exits with code 0. Treat that as the watch ending, not proof that the stack converged; inspect the last object yourself.

​
Read condition history

Use the human-readable table to see every condition reported for each resource:
Each condition includes its status, type, reason, message, and transition age. Start with the newest false or failing condition, then compare its reason and message with runtime logs.
--conditions affects table rendering only in the current CLI. stackdome status --conditions -o json and -o yaml return the same Stack object as ordinary structured status; they do not add condition history.

​
Choose the next diagnostic