View a markdown version of this page

Customer-managed backup and restore - Amazon Timestream
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

For similar capabilities to Amazon Timestream for LiveAnalytics, consider Amazon Timestream for InfluxDB. It offers simplified data ingestion and single-digit millisecond query response times for real-time analytics. Learn more here.

Customer-managed backup and restore

Customer-managed backup and restore lets you create, manage, and restore database backups for your Amazon Timestream for InfluxDB 3 clusters (Core and Enterprise). You can create on-demand backups, configure automated scheduled backups, or enable continuous backups for point-in-time restore—all without opening Amazon support tickets.

Overview

The first backup creates a complete copy of your database. Subsequent backups are incremental and capture only the data that changed since the last backup, reducing storage costs and backup time.

InfluxDB 3 also supports continuous backups, which back up the database in the background at all times, enabling point-in-time restore to any moment within the retention window.

Note

Data written within the last 15 minutes might not be included in the backup. To recover the most recent data, use point-in-time restore with a timestamp at least 15 minutes in the past.

Key concepts

ConceptDescription
DbBackupA backup of your database containing the cluster configuration at the time of backup.
DbBackupConfigurationAutomated recurring backups at a defined frequency and retention.
RestoreCreates a new DB from a DbBackup or restores an existing resource back to the provided DbBackup.
RetentionDaysDefines the number of days a backup is kept before automatic deletion.
On Demand BackupTriggers a one-time backup of the database at a point in time.
Continuous BackupThe database backs up in the background at all times and is available for point-in-time restore within the retention window.
Custom Schedule BackupDefines a custom Amazon cron schedule when the DB will be backed up automatically.

Creating a DbBackup

On-demand backup (Amazon CLI)

Create an on-demand backup of a cluster:

aws timestream-influxdb create-db-backup \ --db-resource-id my-db-cluster-id \ --name my-db-backup-2026-07-07 \ --retention-days 7

On-demand backup (Amazon Web Services Management Console)

  1. Open the Amazon Timestream for InfluxDB console.

  2. Choose your InfluxDB 3 cluster.

  3. Choose Actions, Create backup.

  4. Enter a backup name and retention period.

  5. Choose Create backup.

Configuring automated backups

Automated backups can be enabled at resource creation time or updated afterwards. Only one DbBackupConfiguration is allowed per type per resource, and a maximum of 4 DbBackupConfigurations can be enabled per resource.

Enable automated backups on create (Amazon CLI)

aws timestream-influxdb create-db-cluster \ --name my-db-cluster \ --db-instance-type db.influx.medium \ --vpc-subnet-ids subnet-0123456789abcdef0 \ --vpc-security-group-ids sg-0123456789abcdef0 \ --db-parameter-group-identifier InfluxDBV3Core \ --db-backup-configurations '[{ "enabled": true, "type": "CONTINUOUS", "retentionDays": 30 }, { "enabled": true, "type": "DAILY", "retentionDays": 7 }, { "enabled": true, "type": "WEEKLY", "retentionDays": 30 }, { "enabled": true, "type": "MONTHLY", "retentionDays": 365 }]'

Enable automated backups on update (Amazon CLI)

DbBackupConfigurations can be enabled, disabled, and retention can be changed after creation:

aws timestream-influxdb update-db-cluster \ --db-cluster-id my-db-cluster-id \ --db-backup-configurations '[{ "enabled": true, "type": "CONTINUOUS", "retentionDays": 30 }, { "enabled": true, "type": "DAILY", "retentionDays": 7 }, { "enabled": false, "type": "WEEKLY", "retentionDays": 30 }]'

DbBackupConfiguration types

TypeDescription
CONTINUOUSCreates a continuous backup that can be restored to any point in time within the last 15 minutes up to the creation date or retention date (whichever is later). Maximum retention is 35 days.
HOURLYCreates a DbBackup every hour
DAILYCreates a DbBackup every day at midnight in the timezone of the Region
WEEKLYCreates a DbBackup every week on Sunday at midnight in the timezone of the Region
MONTHLYCreates a DbBackup every month on the 1st at midnight in the timezone of the Region
CUSTOMUse an Amazon Cron Expression for a custom frequency. Must be at a daily frequency or less.

Example CUSTOM type:

--db-backup-configurations '[{ "enabled": true, "type": "CUSTOM", "retentionDays": 7, "customSchedule": "cron(00 18 * * * *)" }]'

This creates a backup that runs at 18:00 every day in the local timezone.

Note

Automated backups can start anytime within an hour of the scheduled time.

Best practices for scheduling

  • Use CONTINUOUS for maximum coverage with point-in-time restore and up to 35-day retention.

  • Use longer frequencies like WEEKLY or MONTHLY with longer retention days for extended archival.

Listing and viewing backups

List all backups for a resource:

aws timestream-influxdb list-db-backups \ --db-resource-id my-resource-id

List all backups in the account:

aws timestream-influxdb list-db-backups

Get backup details:

aws timestream-influxdb get-db-backup \ --identifier my-db-backup-id

DbBackup statuses

StatusDescription
IN_PROGRESSBackup is being created (cluster remains available)
COMPLETEDBackup is completed and available to use for restore
FAILEDBackup creation failed
DELETINGBackup is being deleted
DELETEDBackup has been deleted

Restoring from a DbBackup

A DbBackup can be restored to a new resource, or it can replace the existing resource where the backup was taken.

Restore to a new resource

This is the default and recommended restore mode since it does not impact any existing resources.

aws timestream-influxdb restore-from-db-backup \ --name my-new-restored-database \ --db-backup-id my-backup-id \ --restore-mode NEW_RESOURCE

If no additional parameters are provided, the new resource inherits all parameters from the DbBackup. A subset of these parameters can be overridden during restore if provided.

Point-in-time restore (continuous backups)

For a CONTINUOUS backup, you can specify a --restore-to-time parameter for the exact UNIX epoch timestamp in seconds to restore the database. The timestamp must be between the last 15 minutes up to the retention days or time of backup creation (whichever is later).

aws timestream-influxdb restore-from-db-backup \ --name my-pitr-restored-database \ --db-backup-id my-continuous-backup-id \ --restore-mode NEW_RESOURCE \ --restore-to-time 1751382000

Restore to an existing resource

aws timestream-influxdb restore-from-db-backup \ --name my-existing-resource-name \ --db-backup-id my-backup-id \ --restore-mode REPLACE_EXISTING

When restoring to an existing resource, no additional parameters are accepted and no changes to the existing database configuration will be made besides restoring the data.

Warning

Restoring to an existing resource is a destructive operation. The resource will not be available during this operation. Existing data will be deleted and replaced with the data from the time of the backup. If there are failures during the operation, the service will attempt to roll back the resource with the original data. It is highly recommended to take an on-demand DbBackup prior to performing this action. After the operation completes, the original data is deleted and not recoverable unless a DbBackup was taken.

You can restrict this operation using an IAM policy:

{ "Version": "2012-10-17", "Statement": [ { "Effect": "Deny", "Action": "timestream-influxdb:RestoreFromDbBackup", "Resource": "*", "Condition": { "StringEquals": { "timestream-influxdb:RestoreMode": "REPLACE_EXISTING" } } } ] }

Restore (Amazon Web Services Management Console)

  1. Open the Amazon Timestream for InfluxDB console.

  2. In the navigation pane, choose Backups.

  3. Select the backup you want to restore from.

  4. Choose Actions, Restore backup.

  5. Choose the restore mode (New resource or Replace existing).

  6. For continuous backups, optionally specify the point-in-time to restore to.

  7. Configure settings as needed and choose Restore.

Deleting backups

aws timestream-influxdb delete-db-backup \ --identifier my-db-backup-id

If retentionDays is set, the DbBackup is automatically deleted on that day.

Retaining backups when deleting a resource

When deleting a resource, by default all automated DbBackups are deleted and on-demand backups are retained. To retain automated backups on resource deletion, use the --retain-automated-backups flag:

aws timestream-influxdb delete-db-cluster \ --db-cluster-id my-db-cluster-id \ --retain-automated-backups

Retained automated backups are automatically deleted at their configured expiry day depending on the retention policy.

Performance impact

Backups on Amazon Timestream for InfluxDB v3 do not have any notable impact to the performance of the database.

The first backup may take longer since it creates a full database backup. Subsequent backups are incremental and only capture changed data.

The backup start time can be delayed due to scheduling but should generally start within one hour of the on-demand backup or scheduled time.

Relationship to service-managed backups

With customer-managed backup and restore, you have full control over your backup strategy including continuous backups for point-in-time restore:

FeatureService-managed backupsCustomer-managed backups
ControlManaged by AmazonManaged by you
AccessVia Amazon support ticketSelf-service (Console, CLI, API)
FrequencyAutomaticOn-demand, scheduled, or continuous
Point-in-time restoreNot availableAvailable with CONTINUOUS type (up to 35-day window)
RetentionManaged by AmazonDefined by you (per configuration)
Restore methodSupport ticket requiredSelf-service restore (new resource or replace existing)

Security and encryption

  • All backups are encrypted at rest using service-managed Amazon KMS keys.

  • For resources with customer managed keys (CMK) enabled, backups are encrypted with the same Amazon KMS key.

  • Access to backup operations is controlled through IAM policies.

  • Backup data remains in the same Amazon Web Services Region as your cluster.

Limitations

  • Backups can only be restored within the same Amazon Web Services account and Region where they were created.

  • Cross-Region and cross-account restore is not supported.

  • A maximum of 4 DbBackupConfigurations can be enabled per resource.

  • Custom schedule frequency must be at a daily frequency or less.

  • Continuous backup maximum retention is 35 days.

  • Point-in-time restore is only available for continuous backups, not scheduled backups.

  • In-place restore (REPLACE_EXISTING mode) is not supported for clusters encrypted with a customer managed key (CMK). Use NEW_RESOURCE mode instead.

Pricing

There is no additional charge for the customer-managed backup and restore feature. You pay only for the Amazon S3 storage used by your backups. Incremental backups reduce storage costs by capturing only changed data. For Amazon S3 storage pricing, see Amazon S3 pricing.