Closes #5. The status was read from the retained topic in exactly one place -- init_machine -- so anything that cleared the machine's shared state left the page showing `UNKNOWN` with an empty status text until someone printed or the pods restarted. Two ways to hit that, both seen on the cluster: a manual restart from the Admin Center, and a Redis restart, which drops every `machine:*` key at once because this deployment runs Redis without persistence by design. The framework already offered both hooks and this driver implemented neither: restart_machine called on a manual restart ping_machines called periodically when MACHINE_PING_ENABLED is set, which is on by default and confirmed on here That last one is the real fix -- it is what recovers the page after a cache wipe with nobody printing, which is the case the issue was actually about. The issue text claimed the framework "does not obviously offer a hook" for periodic work; that was wrong, and it is corrected there. ping_machines guards each machine separately so one broker problem does not stop the others refreshing. _refresh_status already turns an unreachable broker into an UNKNOWN status rather than an exception, so the guard is for everything else. print_labels is deliberately unchanged. With AWAIT_RESULT_S=0 the page sits at PRINTING after a dispatch, and the periodic refresh now corrects that within a ping interval -- which is better than reading the retained topic immediately, when the agent has not yet published anything about the job.

Commits

Commits on Jul 30, 2026

fix(status): refresh on restart and periodically, not only at init