Maximo Open Forum

Hi Community,

I'm looking for guidance on handling a specific disaster recovery scenario for IBM Maximo running on OpenShift-particularly when a namespace (e.g., manage) is accidentally deleted.

We've tested backup and restore using both OADP (Velero) and IBM Storage Fusion, along with the standard .sh scripts provided by IBM.

Using OADP:

• The restore process faced several challenges:

  • Failures in ClusterServiceVersions (CSVs) and Subscriptions

  • Operator instances(especially for Maximo Manage) failed to deploy correctly

• Even after successful script execution, restoration was only partially successful.

Using IBM Storage Fusion:

• Restore was mostly successful, and the Maximo application came back online.

• However, we encountered an etcd restore error (BMYBR0009) related to a resource group. IBM documentation suggests escalating this to support with logs.

Despite following IBM's instructions and using the .sh automation scripts:

• Only basic resources (like Secrets) and the namespace (manage) were restored.

• Critical OpenShift components such as:

  • Deployments

  • Pods

  • Subscriptions

  • Operators
    were not restored, leading to an incomplete recovery.

From OpenShift and IBM MAS disaster recovery perspectives:

• What is the recommended approach to fully recover a deleted namespace that includes both Maximo data and all OpenShift-managed objects?

• Should we be maintaining a standby disaster recovery cluster, or can this be handled within the same cluster using backup?