Disaster recovery in Cloud Pak for Business Automation

Learn about the disaster recovery options in IBM Cloud Pak® for Business Automation.

Disaster recovery in Cloud Pak for Business Automation

The Cloud Pak for Business Automation disaster recovery processes consist of the following procedures.

Setting up the disaster recovery backup environment:

In the primary IBM Cloud Pak for Business Automation environment, scale down the replicas to 0.
Redeploy or restart IBM Cloud Pak for Business Automation in your backup environment.
1. Customize the custom resource with the replicated persistent volume (PV) and persistent volume claim (PVC) from the primary environment.
2. Customize the custom resource to point to the Lightweight Directory Access Protocol (LDAP) and replicated databases from the primary environment.
3. If you want to include IBM Automation Decision Services designer or runtime, configure the ConfigMaps that contain the certificates that access Git servers, machine learning providers, and other servers that have untrusted certificates.
Validate that your backup environment can access the backup data.

A failover operation is the process of switching production from your primary environment to your recovery site. If you need to complete a disaster recovery failover:

Find the time of the most recent replication for both your storage and database.
Roll the database forward to align with the most recent storage replication time.
Resync missing content if the database roll forward did not sync with storage.
Scale up your IBM Cloud Pak for Business Automation backup environment.

A failback operation is the process of restoring production to the primary environment after a disaster or a scheduled maintenance period. If you need to complete a disaster recovery failback:

Replicate the database and storage from the backup environment to the primary environment.
Scale down the IBM Cloud Pak for Business Automation replicas to 0 in the backup environment.
Start IBM Cloud Pak for Business Automation in the primary environment.
Configure database and storage replication from the backup environment to the primary environment.

Options for disaster recovery

Private data center with IBM Spectrum Protect Plus

The following image shows a high-level architecture where IBM Spectrum Protect Plus and a private data center is used to provide disaster recovery.

The architecture in the previous image is good for protecting Red Hat® OpenShift® and Kubernetes environments, and consists of a snapshot-based storage backup in addition to continuous incremental backup to vSnap. The following describes the architecture:

A vSnap environment can include copying data to a standard object storage, an object storage archive, or IBM Spectrum Protect storage.
A snapshot-based storage or a vSnap environment can help you recover data.
The unified user interface is used by backup administrators, virtual infrastructure administrators, and application administrators.
This architecture follows a centralized role-based access control system.

The following diagram shows a detailed architecture where IBM Spectrum Protect Plus and a private data center are used to provide disaster recovery. IBM Spectrum Protect Plus is used to back up and restore persistent volume claims (PVC) and persistent volumes (PV). Backup data is synchronized by replicating from the primary vSnap server to a backup vSnap server. For more information about vSnap backup, see Establishing a replication partnership for a vSnap server.

Storage class types include Red Hat OpenShift Data Foundation (previously Red Hat OpenShift Container Storage).

In the secondary environment, the IBM Cloud Pak for Business Automation pods are inactive until a disaster recovery event.

Private data center with IBM Spectrum Protect Plus in a third data center

The following diagram shows a disaster recovery architecture that consists of a private data center where IBM Spectrum Protect Plus is hosted in a third data center:

IBM Spectrum Protect Plus is used to schedule backups for persistent volume claims (PVC) and persistent volumes (PV) in the primary environment.
Backup data is stored in the vSnap server.
When required, the PVC is restored to the secondary Red Hat OpenShift environment.

To learn more about IBM Spectrum Protect Plus, see the following links:

IBM Spectrum Protect Plus uses Red Hat OpenShift APIs for Data Protection to backup and restore Red Hat OpenShift cluster resources. For more information, see For OpenShift: Installing and configuring Velero by using the OpenShift APIs for Data Protection operator.
IBM Spectrum Protect Plus has a container backup support to backup PVCs, cluster-scoped resources, and namespace-scoped resources. For more information, see Installing Container Backup Support.
IBM Spectrum Protect Plus uses vSnap server to backup data. For more information, see Installing a vSnap server.

Private data center with Red Hat OpenShift Container Storage replication

The following diagram shows Cloud Pak for Business Automation in a private data center with Red Hat OpenShift Data Foundation (previously Red Hat OpenShift Container Storage). Disaster recovery involves Red Hat OpenShift Data Foundation replication and database replication from primary to secondary.

FileNet P8 container high availability topology

Multiple zones on IBM Cloud Public

The following diagram shows a disaster recovery environment that consists of multiple zones on IBM Cloud® Public. In Red Hat OpenShift clusters, the nodes are distributed across multiple zones. Cloud Pak for Business AutomationThe FileNet P8 container is deployed with multiple replicas.

Disaster recovery in Automation Decision Services

The following table shows disaster recovery in Cloud Pak for Business Automation for Automation Decision Services.

Table 1. Disaster recovery in Automation Decision Services in Cloud Pak for Business Automation
Data type	IBM Cloud Pak for Business Automation environment
Configuration files	Kubernetes secrets and the IBM Cloud Pak for Business Automation custom resource can be used to back up and restore your environment. Any changes that are made in your primary environment must be replicated to the backup.
Database data (including external MongoDB database)	See Backing up your environment.

Disaster recovery in IBM Business Automation Workflow

The following table shows differences between disaster recovery in Cloud Pak for Business Automation and on-premises environments for IBM Business Automation Workflow.

Table 2. Differences between disaster recovery in IBM Business Automation Workflow in Cloud Pak for Business Automation and in an on-premises environment
Data type	On-premises environment disaster recovery	IBM Cloud Pak for Business Automation environment disaster recovery
Configuration files	There are XML configuration files on your local disk. These files are replicated by the WebSphere® Application Server deployment manager and node agent.	When configmaps are updated, they are also backed up. On the disaster recovery OpenShift Container Platform environment, IBM Cloud Pak for Business Automation imports the default configuration files automatically. You will need to apply the changes for any updated configmaps.
Server logs and other persistent files	The files on your local disk are replicated to the disaster recovery environment according to a schedule.	Files are located on persistent volume claims (PVC) or persistent volumes (PV) which are located in the StorageClass. These files can be replicated through StorageClass capabilities or through a third-party software such as IBM Spectrum Plus.
Transaction logs	Transaction logs are stored in the database.	Transaction logs are stored in the database.
Database data	You can use database replication. For example, Db2® has a high availability and disaster recovery feature for database replication.	You can use database replication. For example, Db2 has a high availability and disaster recovery feature for database replication.
Lucene or Process Federation Server	You need to rebuild the Lucene data in the disaster recovery site.	You need to rebuild the Process Federation Server data in the disaster recovery site.

Disaster recovery in IBM FileNet Content Manager

The following table shows differences between disaster recovery in Cloud Pak for Business Automation and on-premises environments for IBM FileNet® Content Manager.

Table 3. Differences between disaster recovery in IBM FileNet Content Manager in Cloud Pak for Business Automation and in an on-premises environment
Data type	IBM FileNet Content Manager on-premises environment	IBM Cloud Pak for Business Automation environment
Configuration files	Configuration manager profiles and WebSphere Application Server profiles can be used to back up your primary environment. Any changes that are made in your primary environment must be replicated to the backup.	Kubernetes secrets and the IBM Cloud Pak for Business Automation custom resource can be used to back up and restore your environment. Any changes that are made in your primary environment must be replicated to the backup.
Documents in storage	If the files are located in the local storage, create a backup snapshot and copy it to the backup. If the files are located in the cloud storage, enable the storage across availability zones and replicate it into another region.	If the files are located in the local storage, create a backup snapshot and copy it to the backup. If the files are located in the container storage, create a persistent volume claim snapshot and copy it to the backup by using Spectrum Protect Plus. For example, container storages include Red Hat OpenShift Data Foundation. If the files are located in cloud storage, enable the storage across availability zones and replicate it into another region.
Search indexes	Files are stored in the local storage. Create a backup snapshot and copy it to the backup.	If the files are stored in the local storage, create a backup snapshot and copy it to the backup If the files are stored in the container storage, create a persistent volume claim snapshot and copy it to your backup by using Spectrum Protect Plus. For example, the files are stored in Red Hat OpenShift Data Foundation.
Database data (including data in Content Platform Engine and IBM Business Automation Navigator)	You can use online backup and the `SET WRITE SUSPEND` command to align your backup with storage. You can also use database replication. For example, Db2 has a high availability and disaster recovery feature for database replication.	You can use online backup and the `SET WRITE SUSPEND` command to align your backup with storage. You can also use database replication. For example, Db2 has a high availability and disaster recovery feature for database replication.

Disaster recovery in Operational Decision Manager

The following table shows differences between disaster recovery in Cloud Pak for Business Automation and on-premises environments for Operational Decision Manager.

Table 4. Differences between disaster recovery in Operational Decision Manager in Cloud Pak for Business Automation and in an on-premises environment
Data type	Operational Decision Manager on-premises environment	IBM Cloud Pak for Business Automation environment
Configuration files	Configuration manager profiles and WebSphere Application Server profiles can be used to back up your primary environment. Any changes that are made in your primary environment must be replicated to the backup.	Kubernetes secrets and the IBM Cloud Pak for Business Automation custom resource can be used to back up and restore your environment. Any changes that are made in your primary environment must be replicated to the backup.
Database data (including data in IBM Business Automation Navigator)	For Operational Decision Manager on-premise backup, you can: Use online backup and the `SET WRITE SUSPEND` command to align your backup with storage. Use database replication. For example, Db2 has a high availability and disaster recovery feature for database replication. You can replicate your Operational Decision Manager runtime deployment. For information about preparing to replicate your deployment, see Planning for Operational Decision Manager disaster recovery on OpenShift. You can use an external tool to synchronize the data between your primary environment and your backup environment. For example, external tools include IBM Tivoli System Automation for Multiplatforms or Oracle Cloud Infrastructure (OCI) GoldenGate. For more information about setting up high availability and disaster recovery, see Setting up Operational Decision Manager for High Availability Disaster Recovery.	Same as the on-premise environment.