Rolling Restart Procedure ​

To avoid any gap in active coverage, restart nodes sequentially.

1. Restart standbys first, one at a time (the active keeps serving throughout). Confirm each restarted standby reconnects and resumes replication.
2. Finally, bring down the active with Ctrl+C → via handoff (X), the standby is promoted and clients switch over.
3. Bring the old, now-down active back up → it joins as a standby of the current active (there is no automatic failback).