Creating Your First Backup Collection
This guide walks you through creating a backup collection, taking your first backups, and monitoring backup operations.
Creating a backup collection
A backup collection is a schema object stored in the database schema. You create and manage collections using SQL statements, and browse them using schema navigation commands (like ydb scheme ls) since they appear as directories in the database structure.
Create a collection that includes the tables you want to back up together:
-- Create a collection for related tables
CREATE BACKUP COLLECTION production_backups
( TABLE orders
, TABLE products
, TABLE customers
)
WITH ( STORAGE = 'cluster', INCREMENTAL_BACKUP_ENABLED = 'true' );
Taking backups
After creating the collection, take an initial full backup, then use incremental backups for subsequent operations:
-- Take initial full backup
BACKUP production_backups;
-- Make changes to your data...
-- INSERT, UPDATE, or DELETE operations on backed-up tables
-- Later, take incremental backups to capture the changes
BACKUP production_backups INCREMENTAL;
Note
Backup operations run asynchronously and are not idempotent — each BACKUP command creates a new backup. Before retrying after a timeout, check operation status with ydb operation list incbackup.
Monitoring backup operations
Track backup progress and browse your backup structure:
# Check backup operation status
ydb operation list incbackup
# Get details for specific operation
ydb operation get <operation-id>
# Browse backup collections
ydb scheme ls .backups/collections/
# List backups in a collection
ydb scheme ls .backups/collections/production_backups/