High availability for PostgreSQL: standby and failover
What high availability means for a database, how a standby copy and automatic failover shorten an outage, and what Voroa provides for recovery right now.
What this covers
High availability for a database means keeping a second copy running that is continuously kept in step with the first, and switching to it automatically if the primary stops answering. What it buys you is time: recovery becomes a matter of seconds without a human, rather than the time it takes someone to be paged and restore from a backup. It is worth being precise about what it does not buy you. A standby faithfully copies a bad write as well as a good one, so it is not a substitute for backups, and it does not protect you from an accidental delete.
Automatic failover to a standby is not available on Voroa today. Backup and recovery for managed databases are covered below.