wodby cluster gpu-node-pool
Add, resize, replace or remove the cluster's GPU node pool
Add, resize, replace or remove the cluster's optional GPU node pool. Pods that request GPUs run on this pool. Values not given as flags keep those of the current pool. Changing the machine type creates the new pool before the old one is deleted. --remove deletes the pool; it is refused while pods request GPUs.
Usage
wodby cluster gpu-node-pool CLUSTER_ID [flags]Options
| Flag | Description | Default |
|---|---|---|
--data string |
JSON request body | - |
--disk-size int |
Node disk size in GB, at least 20; required when the provider's node disk size is configurable | 0 |
-f, --file string |
Path to JSON request body | - |
-h, --help |
help for gpu-node-pool | false |
--machine-type string |
Machine type that reports GPUs | - |
--max int |
Maximum node count | 0 |
--min int |
Minimum node count; 0 lets the pool scale to no nodes where the provider supports it | 0 |
--remove |
Remove the GPU node pool | false |
--timeout duration |
Maximum time to wait | 10m0s |
--wait |
Wait for the created task or deployment to finish | false |
-y, --yes |
Confirm --remove without prompting | false |
Options inherited from parent commands
| Flag | Description | Default |
|---|---|---|
--access-token string |
Access token | - |
--api-base-url string |
Public REST API base URL | https://apiv2.wodby.com/v1 |
--api-key string |
API key | - |
--ci-config-path string |
Path to CI config | /tmp/.wodby-ci.json |
-o, --output string |
Output format: table, vertical, or json | table |
--verbose |
Verbose output | false |