Mismatch in GPU Node allocatable and node's capacity in RHOCP 4
Issue
- GPU Node's "allocatable nvidia.com/gpu" showing 1 where as node's "capacity nvidia.com/gpu" showing 2
- GPU Node's "allocatable nvidia.com/gpu" reduced to 1
- Difference value in GPU node allocatable and capacity for nvidia.com/gpu
allocatable:
cpu: 35500m
ephemeral-storage: '288342967022'
hugepages-1Gi: '0'
hugepages-2Mi: '0'
memory: 927688156Ki
nvidia.com/gpu: '1'
pods: '250'
capacity:
cpu: '36'
ephemeral-storage: 314037228Ki
hugepages-1Gi: '0'
hugepages-2Mi: '0'
memory: 928839132Ki
nvidia.com/gpu: '2'
pods: '250'
Environment
- Red Hat OpenShift Container Platform (RHOCP)
- 4
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.