kubernetes
/
kops
mirror of https://github.com/kubernetes/kops.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix no-schedule issue for the GPU operator

This commit is contained in:
Christian Schlichtherle 2021-02-25 11:28:22 +01:00 committed by Ciprian Hacman
parent 7759c754cb
commit d196efcb82
1 changed files with 18 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
docs/gpu.md
View File

@ -41,22 +41,37 @@ In order to match the kOps environment, create a `values.yaml` file with the fol
operator:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
driver:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
toolkit:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
devicePlugin:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
dcgmExporter:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
gfd:
nodeSelector:
@ -64,12 +79,14 @@ gfd:
tolerations:
- key: nvidia.com/gpu
operator: Exists
effect: NoSchedule
node-feature-discovery:
worker:
nodeSelector:
kops.k8s.io/instancegroup: gpu-nodes
tolerations:
- key: nvidia.com/gpu
operator: Exists
```
Once you have installed the the _helm chart_ you should be able to see the GPU operator resources being spawned in the `gpu-operator-resources` namespace.