I've created new worker nodes that are dedicated to running Longhorn replicas. These nodes are tainted with the `node-role.kubernetes.io/longhorn` taint, so no regular pods will be scheduled there by default. Longhorn pods thus needs to be configured to tolerate that taint, and to be scheduled on nodes with the similarly-named label.
7 lines
201 B
YAML
7 lines
201 B
YAML
apiVersion: longhorn.io/v1beta2
|
|
kind: Setting
|
|
metadata:
|
|
name: taint-toleration
|
|
namespace: longhorn-system
|
|
value: du5t1n.me/machine=raspberrypi:NoExecute;node-role.kubernetes.io/longhorn:NoSchedule
|