imagePullSecrets
Image pull secrets to configure if using custom private images. | list | |
nvidia-device-plugin
Options for the Coreweave fork of the Nvidia device plugin chart. This chart builds on the default configuration provided by Nvidia, and uses these default chart values. | object | See default chart values. |
operator.affinity
The affinity for the operator deployment. | object | nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: node.coreweave.cloud/class
operator: In
values:
- cpu
|
operator.config.operator.nodeSet.failedPodsBackoffGCInterval
The time that has to pass before next iteration of backoff GC is run for checking failed pods. | string | |
operator.config.operator.nodeSet.forceScalingDeleteKnownConditionTimeout
The delay to wait before deleting a pod during nodeset scaling with a known Slurm state. "0" disables this feature. | string | |
operator.config.operator.nodeSet.forceScalingDeleteUnknownConditionTimeout
The delay to wait before deleting a pod during nodeset scaling with a unknown Slurm state. "0" disables this feature. | string | |
operator.config.operator.nodeSet.maxBurstReplicas
A rate limiter for booting pods when there are a lot of pods. A too high of a value can cause registry DoS issues. | int | |
operator.config.operator.nodeSet.statusUpdateBackoffGCInterval
The time that has to pass before next iteration of backoff GC is run for checking node status updates. | string | |
operator.config.operator.nodeSlice.maxNodesPerNodeSlice
The maximum number of nodes that can be in a single nodeSlice. | int | |
operator.image
The image to use for the operator. | object | repository: registry.gitlab.com/coreweave/sunk/operator
tag:
|
operator.leaderElection.enabled
This forces the operator to use leader election even if the number of replicas is set to 1.
Useful if planning to scale after deployment. | bool | |
operator.leaderElection.leaderElectionID
The string value to use as the leader election id. | string | |
operator.logLevel
The log level.
Uses integers or zap log level strings: debuginfowarnerrordpanicpanicfatal | string | |
operator.maxConcurrentReconciles
| int | |
operator.podMonitor.enabled
Enable monitoring via the Prometheus operator PodMonitor CRD. | bool | |
operator.priorityClassName
The priority class name for the operator. | string | |
operator.replicas
The number of replicas of the operator pod to run.
Leader election will be enabled if this is greater than 1 or leader election is explicitly enabled. | int | |
operator.resources
The resource to request for the operator. | object | limits:
memory: 32Gi
cpu: 16
requests:
cpu: 8
memory: 32Gi
|
operator.tolerations
The tolerations for the operator deployment. | list | |
operator.vmPodScrape.enabled
Enable monitoring via the VictoriaMetrics operator VMPodScrape CRD. Note: To enable this, podMonitor must be disabled in addition to setting vmPodScrape.enabled to true. | bool | |
priorityClass.enabled
Enable the priority class for the control plane components. | bool | |
priorityClass.value
The value of the priority class, generally should be high relative to other priority classes as these are critical components. | int | |
scheduler.podMonitor.enabled
Enable monitoring via the Prometheus operator PodMonitor CRD. | bool | |
scheduler.vmPodScrape.enabled
Enable monitoring via the VictoriaMetrics operator VMPodScrape CRD. Note: To enable this, podMonitor must be disabled in addition to setting vmPodScrape.enabled to true. | bool | |
syncer.podMonitor.enabled
Enable monitoring via the Prometheus operator PodMonitor CRD. | bool | |
syncer.vmPodScrape.enabled
Enable monitoring via the VictoriaMetrics operator VMPodScrape CRD. Note: To enable this, podMonitor must be disabled in addition to setting vmPodScrape.enabled to true. | bool | |
