ductus handover
Hand the running engine over to the binary now on disk and the project as it is now, refusing no connection (Linux: the package's systemctl reload ductus runs it): an upgrade or a project deploy. The new engine starts beside the running one on the same listening sockets; once it's ready, the running one stops and lets go. Exit 0 once the new engine sends; 1 when it couldn't take over (the running engine goes on: the reason is printed).
ductus handover [OPTIONS]
Options
| What it does | Default | Environment | |
|---|---|---|---|
--control <CONTROL> | The running engine's control socket (default: $RUNTIME_DIRECTORY/control, which systemd sets for the service and its reload, else /run/ductus/control). | DUCTUS_CONTROL | |
--timeout <TIMEOUT> | How long the new engine may take to be ready: it migrates the database first. | 5m | |
--with-environment | Start the new engine with this command's environment rather than the running engine's: under systemd, ductus.env as it is now (the unit's reload passes this). |