Skip to main content
Amazon FSx for NetApp ONTAP

Replicate data to FSx for ONTAP in NetApp Workload Factory

Contributors netapp-rlithman netapp-mwallis
PDFs

Create a replication relationship for an FSx for ONTAP file system in NetApp Workload Factory to avoid data loss in case of an unforeseen disaster. You can replicate data between two FSx for ONTAP file systems, or between an on-premises ONTAP system and an FSx for ONTAP file system.

When replicating storage VM data and configuration settings, you must first create and initialize the replication relationship and then wait for the data to sync at least once before completing the cut over operation.

About this task

Replication protects your data if a disaster affects your region; it can also be used for migration purposes.

Replication includes creating and initializing a replication relationship. To set up replication, create the replication relationship. To transfer the data between the source and target storage systems, initialize the replication relationship. When migrating storage VM data and configuration settings, you must also complete the cutover operation.

Replicated volumes in the target file system are data protection (DP) volumes and follow the naming format: {OriginalVolumeName}_copy.

If you replicate a source volume with immutable files, the target volume and file system remain locked until the source volume's retention period ends. The immutable files feature is available when you create a volume for an FSx for ONTAP file system.

Note

You can replicate one source (read/write) volume or one data protection (DP) volume. Cascading replication is supported, but a third hop isn't. Learn more about cascading replication.

Migration use cases

For migration, you can choose to replicate storage VM data and configuration settings for one storage VM. If you migrate data and configuration settings together, make sure the last replication for the volume finished within the last 24 hours. You must select all volumes in the same storage VM to use this feature. The tiering policy for all volumes defaults to the tiering policy of the source volume, which is recommended for migration use cases.

Workload Factory supports migration replication for storage VMs between the following storage systems.

  • On-premises ONTAP systems and FSx for ONTAP file systems

  • Cloud Volumes ONTAP and FSx for ONTAP file systems

  • FSx for ONTAP and FSx for ONTAP file systems

    • First to first generation

    • Second to second generation

To migrate storage VM data and configuration settings, you must complete three operations.

  1. Create the replication relationship, select Migration as the use case and select Replicate storage VM configuration.

  2. Initialize the replication relationship to transfer the data from the source to the target storage system.

  3. Cut over replication for migration use cases to permanently migrate data and configuration settings from the source file system to the target FSx for ONTAP file system.

Step 1: Create the replication relationship

Replicate data between two FSx for ONTAP file systems, or between an on-premises ONTAP system and an FSx for ONTAP file system.

Before you begin

Review these requirements before you begin.

  • Ensure that one FSx for ONTAP file system is available to use for the target in the replication relationship.

  • The FSx for ONTAP file system you use for the replication relationship must have an associated link. Learn how to associate an existing link or to create and associate a new link. After you associate the link, return to this operation.

  • For replication from an on-premises ONTAP system to an FSx for ONTAP file system, make sure you have discovered the on-premises ONTAP system.

  • Replication isn't supported for volumes in a state other than available, created, or misconfigured, and when the ONTAP version isn't compatible.

  • For migration use cases ensure that the last replication for the volume completed in the last 24 hours before you create a replication relationship with storage VM data and configuration settings.

Steps

  1. Log in using one of the console experiences.

  2. Select the menu The hamburger menu icon is used to navigate to workloads like storage and then select Storage.

  3. From the Storage menu, select FSx for ONTAP.

  4. From FSx for ONTAP, select the file system that contains the volume(s) to replicate.

  5. Either replicate all volumes in a file system or replicate select volumes.

    • To replicate all volumes in a file system: From the file system overview, select Replicate data.

    • To replicate select volumes: From the file system overview, select the Volumes tab.

      In the Volumes table, select one or more volumes and then select Replicate data.

  6. On the Replicate data page, under Replication target, provide the following:

    1. Use case: Select one of the following use cases for the replication. Depending on the selected use case, Workload Factory fills in the form with recommended values in accordance with best practices. You can accept the recommended values or make changes as you complete the form.

      • Artificial Intelligence (AI) and machine learning (ML): replicate your on-premises ONTAP data to FSx for ONTAP and create S3 access points to support AI and ML training in the cloud

      • Migration: transfers your data to the target FSx for ONTAP file system

        Replicate storage VM configuration: Optionally, select to replicate storage VM data and configuration settings for a single storage VM. When migrating data and configuration settings simultaneously, ensure that the last replication for the volume completed in the last 24 hours. All volumes in the same storage VM must be selected to use this feature. The tiering policy for all volumes defaults to the tiering policy of the source volume, which is recommended for migration use cases.

      • Hot disaster recovery: ensures high availability and rapid disaster recovery for critical workloads

      • Cold or archive disaster recovery:

        • Cold disaster recovery: uses longer recovery time objectives (RTO) and recovery point objects (RPO) to lower costs

        • Archive: replicates data for long-term storage and compliance

      • Other

        Additionally, the use case selection determines the replication policy, or SnapMirror policy (ONTAP). The terms used to describe replication policies come from ONTAP 9 documentation.

        • For migration and other, the replication policy is called MirrorAllSnapshots. MirrorAllSnapshots is an asynchronous policy for mirroring all snapshots and the latest active file system.

        • For hot, cold, or archive disaster recovery, the replication policy is called MirrorAndVault. MirrorAndVault is an asynchronous and vault policy for mirroring the latest active file system and daily and weekly snapshots.

          For all use cases, if you enable snapshots for long-term retention, the default replication policy is MirrorAndVault.

    2. FSx for ONTAP file system: Select credentials, region, and FSx for ONTAP file system name for the target FSx for ONTAP file system.

    3. Storage VM name: Select the storage VM from the dropdown menu. The storage VM you select is the target for all selected volumes in this replication relationship.

    4. Volume name: The target volume name is generated automatically with the following format {OriginalVolumeName}_copy. You can use the auto-generated volume name or enter another volume name.

    5. Tiering policy: Select the tiering policy for the data stored in the target volume. The tiering policy defaults to the recommended tiering policy for the use case you selected.

      Balanced (Auto) is the default tiering policy when creating a volume using the Workload Factory console. For more information about volume tiering policies, refer to Volume storage capacity in AWS FSx for NetApp ONTAP documentation. Note that Workload factory uses use-case based names in the Workload Factory console for tiering policies and includes FSx for ONTAP tiering policy names in parentheses.

      If you selected the migration use case, Workload Factory automatically selects to copy the tiering policy of source volume to the target volume. You can deselect to copy the tiering policy and select a tiering policy which applies to the volume selected for replication.

    6. Max transfer rate: Select Limited and enter the max transfer limit in MB/s. Alternatively, select Unlimited.

      Without a limit, network and application performance may decline. Alternatively, we recommend an unlimited transfer rate for FSx for ONTAP file systems for critical workloads, for example, those that are used primarily for disaster recovery.

  7. Under Replication settings, provide the following:

    1. Replication interval: Select the frequency that snapshots are transferred from the source volume to the target volume.

    2. Long-term retention: Optionally, enable snapshots for long-term retention. Long-term retention enables business services to continue operating even through a complete site failure, supporting applications to fail over transparently using a secondary copy.

      Replications without long-term retention use the MirrorAllSnapshots policy. Enabling long-term retention assigns the MirrorAndVault policy to the replication.

      If you enable long-term retention, then select an existing policy or create a new policy to define the snapshots to replicate and the number to retain.

      Note Matching source and target labels are required for long-term retention. If desired, Workload factory can create missing labels for you.

      • Choose an existing policy: select an existing policy from the dropdown menu.

      • Create a new policy: enter a policy name.

    3. Immutable snapshots: Optional. Select Enable immutable snapshots to prevent snapshots taken in this policy from being deleted during the retention period.

      • Set the Retention period in number of hours, days, months, or years.

      • Snapshot policies: In the table, select the snapshot policy frequency and the number of copies to retain. You can select more than one snapshot policy.

    4. S3 access point: Optionally, attach an S3 access point to access FSx for ONTAP file system data residing on NFS or SMB/CIFS volumes via AWS S3 APIs. Only the file access type is supported. Providing the following details:

      • S3 access point name: Enter the name of the S3 access point.

      • User: Select an existing user with access to the volume or create a new user.

      • User type: Select UNIX or Windows as the user type.

      • Network configuration: Select Internet or Virtual private cloud (VPC). The type of network you choose determines whether the access point is accessible from the internet or restricted to a specific VPC.

      • Enable inventory table: When you enable the inventory table on the volume, the system generates metadata for all objects accessible to the S3 access point and incurs AWS S3 request costs. Refer to Amazon S3 pricing documentation for more information.

    5. S3 access point tags: Optionally, you can add up to 50 tags.

  1. Select Create.

Result

The replication relationship appears in the Replication relationships tab in the target FSx for ONTAP file system.

If you created a replication relationship for migration purposes, you must cut over all the volumes and their associated storage VM to complete migration of storage VM data and configuration settings to the target FSx for ONTAP file system.

Step 2: Initialize the replication relationship

Initialize a replication relationship between source and target volumes to transfer the snapshot and all data blocks in NetApp Workload Factory.

Initialization performs a baseline transfer: it makes a snapshot of the source volume, then transfers the snapshot and all the data blocks it references to the target volume.

Before you begin

Consider when you choose to complete this operation. Initialization can be time-consuming. You might want to run the baseline transfer in off-peak hours.

Steps

  1. Log in using one of the console experiences.

  2. Select the menu The hamburger menu icon is used to navigate to workloads like storage and then select Storage.

  3. From the Storage menu, select FSx for ONTAP.

  4. From FSx for ONTAP, select the actions menu of the file system to update and then select Manage.

  5. From the file system overview, select the Replication relationships tab.

  6. In the Replication relationships tab, select the actions menu of the replication relationship to initialize.

  7. Select Initialize.

  8. In the Initialize relationship dialog, select Initialize.

    When the replication status shows Idle, the data transfer is complete.

Step 3: Cut over replication for migration use cases

After creating the replication relationship for a migration use case, you must cut over replication to complete migration of storage VM data and configuration settings to the target FSx for ONTAP file system. Cutover replication permanently migrates data and storage VM configuration settings from the source file system to the target FSx for ONTAP file system. During the cutover, data gets replicated for the last time. The source storage VM goes offline when the cutover begins.

Plan for the cutover operation

Consider the following when planning for the cutover operation for migration use cases.

  • Replication version support: Source and target storage systems must have the same major ONTAP version. For example, if the source system is running ONTAP version 9.18.1, then the target system must also run a version of 9.18.1.

  • Storage VM configuration settings: The cutover operation doesn't migrate all storage VM configuration settings to the target storage VM. For example, iSCSI, broadcast-domain MTU, and DNS settings do not migrate. Update these settings manually on the target storage VM after cutover. For more details, refer to the Update configuration settings after cutover section.

  • Clones: Clones get migrated as individual volumes, and therefore, will consume more space on the target FSx for ONTAP file system. Consider deleting the clones before cutover to avoid increased capacity on the target.

  • Cache volumes: For ONTAP FlexCache 9.17.x or below. Source storage VMs that contain cache volumes aren't supported for cutover. Remove the ONTAP FlexCache relationship before cutover to avoid migration failures.

  • AWS S3 access points: Cutover replication for migration use cases isn't supported for volumes with S3 access points. Delete S3 access points from the source volume to avoid migration failures.

  • ONTAP S3 configurations: Cutover replication for migration use cases isn't supported for vservers with ONTAP S3 configurations. Remove ONTAP S3 configurations including object-store-server, buckets, users, groups, or policies on the vserver, to avoid migration failures.

  • LUN serial numbers and iSCSI target IQN: LUN serial numbers and the iSCSI target IQN change during cutover.

Before you begin

Review these requirements before you begin.

  • Stop any client access to your storage VM before you cut over replication.

  • Ensure all source volumes are not serving any data before you cut over replication.

  • Ensure the data is synced between the source and target volumes before you cut over replication.

  • The FSx for ONTAP file system you use for the replication relationship must have an associated link. Learn how to associate an existing link or to create and associate a new link. After you associate the link, return to this operation.

    • The following ports must be open in the security group associated with the FSx for ONTAP file system for link connectivity: port 443 (HTTPS) and port 22 (SSH).

Steps

  1. Log in using one of the console experiences.

  2. Select the menu The hamburger menu icon is used to navigate to workloads like storage and then select Storage.

  3. From the Storage menu, select FSx for ONTAP.

  4. From FSx for ONTAP, select the file system that contains the volume(s) to replicate.

  5. Select the Replication relationships tab.

  6. In the Replication relationships table, select the replication relationship to cut over and then select Cut over replication.

  7. Review the information in the Cut over replication dialog and then type cut over to confirm.

  1. Select Cut over.

Result

After initiating cutover, the source storage VM goes offline.

Update configuration settings after cutover

After you cut over replication for migration use cases, update the following storage VM configuration settings on the target storage VM.

  • For iSCSI:

    • Recreate the igroup

    • Update LUN mapping

    • Log in to each SAN client and update iSCSI/MPIO settings for the new target and remove the old vserver target IPs

  • Validate broadcast-domain MTU settings are as expected

  • Validate DNS settings are as expected

  • Recreate cache relationships (ONTAP FlexCache relationships) if there were cache volumes in the source storage VM.

Modify the tiering policy for the target volume(s) after cutover.