Database Backup and Point-in-Time Recovery with RDS

Intermediate15 min

Create manual RDS snapshots, configure automated backups, and perform point-in-time recovery to restore databases to any second within the retention window.

Prerequisites

-AWS CLI v2 installed and configured
-Existing RDS database instance
-IAM permissions for RDS snapshot operations

Steps

Create a manual snapshot

Take an immediate snapshot of the RDS instance to capture the current state of the database before making changes.

$ aws rds create-db-snapshot --db-instance-identifier my-production-db --db-snapshot-identifier my-production-db-pre-migration-$(date +%Y%m%d)

Always create a manual snapshot before schema migrations or major updates. Manual snapshots persist until you explicitly delete them, unlike automated backups.

Wait for the snapshot to complete

Monitor the snapshot creation progress and wait until it reaches the available state.

$ aws rds wait db-snapshot-available --db-snapshot-identifier my-production-db-pre-migration-$(date +%Y%m%d) && echo 'Snapshot ready'

Snapshot creation time depends on database size and can take from minutes to hours for large databases. The database remains available during snapshot creation but may experience brief I/O suspension.

Check the restore window for point-in-time recovery

Determine the earliest and latest times you can restore to using the automated backup retention window.

$ aws rds describe-db-instances --db-instance-identifier my-production-db --query 'DBInstances[0].{LatestRestorableTime:LatestRestorableTime,BackupRetentionPeriod:BackupRetentionPeriod,BackupWindow:PreferredBackupWindow}' --output table

LatestRestorableTime is typically within the last 5 minutes. The restore window extends back by the number of days in BackupRetentionPeriod.

Restore to a point in time

Create a new RDS instance restored to a specific point in time. This creates a completely new database instance, leaving the original untouched.

$ aws rds restore-db-instance-to-point-in-time --source-db-instance-identifier my-production-db --target-db-instance-identifier my-production-db-restored --restore-time 2026-03-11T08:30:00Z --db-instance-class db.t3.medium --vpc-security-group-ids sg-0example

Point-in-time restore creates a new instance with a new endpoint. You must update your application's connection string to point to the restored instance.

Monitor restore progress

Wait for the restored instance to become available and check its status.

$ aws rds wait db-instance-available --db-instance-identifier my-production-db-restored && aws rds describe-db-instances --db-instance-identifier my-production-db-restored --query 'DBInstances[0].{Status:DBInstanceStatus,Endpoint:Endpoint.Address,Port:Endpoint.Port}' --output table

Clean up old snapshots

List and delete snapshots older than your retention policy to manage storage costs.

$ aws rds describe-db-snapshots --db-instance-identifier my-production-db --snapshot-type manual --query 'DBSnapshots | sort_by(@, &SnapshotCreateTime) | [*].{Snapshot:DBSnapshotIdentifier,Created:SnapshotCreateTime,Size:AllocatedStorage}' --output table

Automated snapshots are free up to your allocated storage size. Manual snapshots incur standard EBS snapshot storage charges at approximately $0.095/GB-month.

Full Script

full-script.sh

#!/bin/bash
# RDS Snapshot and Point-in-Time Recovery
# Usage: ./rds-backup.sh <action> <db-identifier> [restore-time]
# Actions: snapshot, restore, list, cleanup

set -euo pipefail

ACTION="${1:?Usage: $0 <snapshot|restore|list|cleanup> <db-identifier> [restore-time]}"
DB_ID="${2:?Provide RDS instance identifier}"

case "$ACTION" in
  snapshot)
    SNAPSHOT_ID="$DB_ID-manual-$(date +%Y%m%d-%H%M%S)"
    echo "Creating snapshot: $SNAPSHOT_ID"
    aws rds create-db-snapshot \
      --db-instance-identifier "$DB_ID" \
      --db-snapshot-identifier "$SNAPSHOT_ID"

    echo "Waiting for snapshot to complete..."
    aws rds wait db-snapshot-available --db-snapshot-identifier "$SNAPSHOT_ID"
    echo "Snapshot complete: $SNAPSHOT_ID"
    ;;

  restore)
    RESTORE_TIME="${3:?Provide restore time in ISO 8601 format (e.g., 2026-03-11T08:30:00Z)}"
    TARGET_ID="$DB_ID-restored-$(date +%s)"
    echo "Restoring $DB_ID to $RESTORE_TIME as $TARGET_ID ..."

    aws rds restore-db-instance-to-point-in-time \
      --source-db-instance-identifier "$DB_ID" \
      --target-db-instance-identifier "$TARGET_ID" \
      --restore-time "$RESTORE_TIME"

    echo "Waiting for restored instance..."
    aws rds wait db-instance-available --db-instance-identifier "$TARGET_ID"

    ENDPOINT=$(aws rds describe-db-instances \
      --db-instance-identifier "$TARGET_ID" \
      --query 'DBInstances[0].Endpoint.Address' --output text)
    echo "Restored instance available at: $ENDPOINT"
    ;;

  list)
    echo "=== Automated Backups ==="
    aws rds describe-db-instances --db-instance-identifier "$DB_ID" \
      --query 'DBInstances[0].{LatestRestore:LatestRestorableTime,RetentionDays:BackupRetentionPeriod}' \
      --output table

    echo ""
    echo "=== Manual Snapshots ==="
    aws rds describe-db-snapshots \
      --db-instance-identifier "$DB_ID" \
      --snapshot-type manual \
      --query 'DBSnapshots | sort_by(@, &SnapshotCreateTime) | reverse(@) | [*].{Snapshot:DBSnapshotIdentifier,Created:SnapshotCreateTime,SizeGB:AllocatedStorage,Status:Status}' \
      --output table
    ;;

  cleanup)
    DAYS="${3:-30}"
    CUTOFF=$(date -d "$DAYS days ago" +%Y-%m-%dT%H:%M:%S)
    echo "Finding manual snapshots older than $DAYS days (before $CUTOFF)..."

    aws rds describe-db-snapshots \
      --db-instance-identifier "$DB_ID" \
      --snapshot-type manual \
      --query "DBSnapshots[?SnapshotCreateTime<'$CUTOFF'].DBSnapshotIdentifier" \
      --output text | tr '\t' '\n' | while read -r SNAP; do
        echo "Deleting: $SNAP"
        aws rds delete-db-snapshot --db-snapshot-identifier "$SNAP"
      done
    echo "Cleanup complete."
    ;;

  *)
    echo "Unknown action: $ACTION (use snapshot, restore, list, or cleanup)"
    exit 1
    ;;
esac

FAQ

Discussion

Loading comments...

#!/bin/bash # RDS Snapshot and Point-in-Time Recovery # Usage: ./rds-backup.sh <action> <db-identifier> [restore-time] # Actions: snapshot, restore, list, cleanup set -euo pipefail ACTION="${1:?Usage: $0 <snapshot|restore|list|cleanup> <db-identifier> [restore-time]}" DB_ID="${2:?Provide RDS instance identifier}" case "$ACTION" in snapshot) SNAPSHOT_ID="$DB_ID-manual-$(date +%Y%m%d-%H%M%S)" echo "Creating snapshot: $SNAPSHOT_ID" aws rds create-db-snapshot \ --db-instance-identifier "$DB_ID" \ --db-snapshot-identifier "$SNAPSHOT_ID" echo "Waiting for snapshot to complete..." aws rds wait db-snapshot-available --db-snapshot-identifier "$SNAPSHOT_ID" echo "Snapshot complete: $SNAPSHOT_ID" ;; restore) RESTORE_TIME="${3:?Provide restore time in ISO 8601 format (e.g., 2026-03-11T08:30:00Z)}" TARGET_ID="$DB_ID-restored-$(date +%s)" echo "Restoring $DB_ID to $RESTORE_TIME as $TARGET_ID ..." aws rds restore-db-instance-to-point-in-time \ --source-db-instance-identifier "$DB_ID" \ --target-db-instance-identifier "$TARGET_ID" \ --restore-time "$RESTORE_TIME" echo "Waiting for restored instance..." aws rds wait db-instance-available --db-instance-identifier "$TARGET_ID" ENDPOINT=$(aws rds describe-db-instances \ --db-instance-identifier "$TARGET_ID" \ --query 'DBInstances[0].Endpoint.Address' --output text) echo "Restored instance available at: $ENDPOINT" ;; list) echo "=== Automated Backups ===" aws rds describe-db-instances --db-instance-identifier "$DB_ID" \ --query 'DBInstances[0].{LatestRestore:LatestRestorableTime,RetentionDays:BackupRetentionPeriod}' \ --output table echo "" echo "=== Manual Snapshots ===" aws rds describe-db-snapshots \ --db-instance-identifier "$DB_ID" \ --snapshot-type manual \ --query 'DBSnapshots | sort_by(@, &SnapshotCreateTime) | reverse(@) | [*].{Snapshot:DBSnapshotIdentifier,Created:SnapshotCreateTime,SizeGB:AllocatedStorage,Status:Status}' \ --output table ;; cleanup) DAYS="${3:-30}" CUTOFF=$(date -d "$DAYS days ago" +%Y-%m-%dT%H:%M:%S) echo "Finding manual snapshots older than $DAYS days (before $CUTOFF)..." aws rds describe-db-snapshots \ --db-instance-identifier "$DB_ID" \ --snapshot-type manual \ --query "DBSnapshots[?SnapshotCreateTime<'$CUTOFF'].DBSnapshotIdentifier" \ --output text | tr '\t' '\n' | while read -r SNAP; do echo "Deleting: $SNAP" aws rds delete-db-snapshot --db-snapshot-identifier "$SNAP" done echo "Cleanup complete." ;; *) echo "Unknown action: $ACTION (use snapshot, restore, list, or cleanup)" exit 1 ;; esac