Describe the issue: OCS nodes cannot be rolling upgraded because of short of resource, drain blocked due to PDB (PodDisruptionBudget) for OSD pods. The OCS nodes are 10 vCPU 32GB mem * 3 nodes = 30 vCPU and 96GB memory total. During the rolling upgrade, one OCS node stuck at the drain phase. The osd pods are stuck at the Pending state due to insufficient memory. So it looks like the "Resource requirements" doesn't capture the resource requirements during the rolling update. It needs more memory and possibly vCPU for the drain scenario, but the doc doesn't mention the spare resource for the upgrade scenario. Suggestions for improvement: Describe resource requirements for drain scenario. Clarify if we need to meet the requirement when 1 node got offline. Document URL: https://access.redhat.com/documentation/en-US/red_hat_openshift_container_storage/4.6/html-single/planning_your_deployment/index#resource-requirements_rhocs