blakeshome-charts icon indicating copy to clipboard operation
blakeshome-charts copied to clipboard

Disable heathchecks for old NVIDIA GPU

Open phcollignon opened this issue 10 months ago • 0 comments

Some old NVIDIA GPU might be to old to be healtchecked, that results in an error "0/1 nodes are available: 1 Insufficient nvidia.com/gpu" A workaround is to disable XID errors health checks with DP_DISABLE_HEALTHCHECKS=xids ENV property. I added a helm value to optionally add this ENV property to the Pod.

phcollignon avatar Apr 23 '24 18:04 phcollignon