Support for configurable PodDisruptionBudgets (PDB) for CACC

We request full, documented support for Kubernetes PodDisruptionBudgets (PDBs) in IBM Red Hat OpenShift environments running Cognos Analytics Certified Container (CACC), to protect Cognos availability during planned operations (node drain, cluster upgrades, etc.).

PDBs control how many pods of a workload can be disrupted at once using:

• minAvailable: minimum number of pods that must remain running

• maxUnavailable: maximum number of pods that may be disrupted concurrently

Problem

Without proper PDB support and guidance for CACC:

• Planned maintenance (node drain, cluster upgrades) can take too many Cognos pods down simultaneously, causing service disruption.

• We cannot formally express and enforce availability requirements for critical Cognos components.

• Operations must work around this manually, increasing risk and effort.

Requested capabilities

PDB support and enforcement

• Ability to define PDBs for CACC deployments (and their underlying Deployments/StatefulSets).

• Support minAvailable and maxUnavailable (absolute and percentage).

• Ensure OpenShift/cluster operations initiated by IBM (node drain, upgrades, autoscaling) respect these PDBs.

CACC-specific integration

• Clear guidance which Cognos services/components should have PDBs and with which recommended values.

• Validation or examples to avoid impossible settings (e.g. minAvailable > replicas) in typical CACC topologies.

Documentation & best practices

• Official documentation for using PDBs with CACC on IBM Red Hat OpenShift.

• Best-practice examples for:

  • Highly available Cognos environments

  • Non-critical/batch Cognos workloads

Business value

• Increased availability and fewer Cognos outages during planned maintenance.

• More predictable, safer upgrades and operations.

• Better alignment with SLAs and enterprise availability requirements for Cognos Analytics.