118 KiB
| layout | parent | title |
|---|---|---|
| page | CRD reference | ArangoMLExtension V1Alpha1 |
API Reference for ArangoMLExtension V1Alpha1
Spec
.spec.deployment.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.deployment.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.deployment.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.deployment.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.deployment.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.deployment.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.deployment.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.deployment.gpu
Type: boolean [ref]
GPU defined if GPU Jobs are enabled.
Default Value: false
.spec.deployment.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.deployment.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.deployment.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.deployment.image
Type: string [ref]
Image define image details
.spec.deployment.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.deployment.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.deployment.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.deployment.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.deployment.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.deployment.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.deployment.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.deployment.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.deployment.port
Type: integer [ref]
Port defines on which port the container will be listening for connections
.spec.deployment.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.deployment.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.deployment.replicas
Type: integer [ref]
Replicas defines the number of replicas running specified components. No replicas created if no components are defined.
Default Value: 1
.spec.deployment.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.deployment.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.deployment.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.deployment.service.type
Type: string [ref]
Type determines how the Service is exposed
Links:
Possible Values:
"ClusterIP"(default) - service will only be accessible inside the cluster, via the cluster IP"NodePort"- service will be exposed on one port of every node, in addition to 'ClusterIP' type"LoadBalancer"- service will be exposed via an external load balancer (if the cloud provider supports it), in addition to 'NodePort' type"ExternalName"- service consists of only a reference to an external name that kubedns or equivalent will return as a CNAME record, with no exposing or proxying of any pods involved"None"- service is not created
.spec.deployment.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.deployment.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.deployment.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.deployment.tls.altNames
Type: array [ref]
AltNames define TLS AltNames used when TLS on the ArangoDB is enabled
.spec.deployment.tls.enabled
Type: boolean [ref]
Enabled define if TLS Should be enabled. If is not set then default is taken from ArangoDeployment settings
.spec.deployment.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.deployment.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.deployment.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.deployment.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.init.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.init.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.init.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.init.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.init.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.init.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.init.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.init.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.init.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.init.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.init.image
Type: string [ref]
Image define image details
.spec.init.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.init.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.init.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.init.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.init.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.init.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.init.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.init.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.init.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.init.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.init.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.init.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.init.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.init.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.init.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.init.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.init.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.init.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.init.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.init.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.featurization.cpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.featurization.cpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.featurization.cpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.featurization.cpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.featurization.cpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.featurization.cpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.featurization.cpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.featurization.cpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.featurization.cpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.featurization.cpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.featurization.cpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.featurization.cpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.featurization.cpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.featurization.cpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.featurization.cpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.featurization.cpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.featurization.cpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.featurization.cpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.featurization.cpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.featurization.cpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.featurization.cpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.featurization.cpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.featurization.cpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.featurization.cpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.featurization.cpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.featurization.cpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.featurization.cpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.featurization.cpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.featurization.cpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.featurization.cpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.featurization.cpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.featurization.gpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.featurization.gpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.featurization.gpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.featurization.gpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.featurization.gpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.featurization.gpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.featurization.gpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.featurization.gpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.featurization.gpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.featurization.gpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.featurization.gpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.featurization.gpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.featurization.gpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.featurization.gpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.featurization.gpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.featurization.gpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.featurization.gpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.featurization.gpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.featurization.gpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.featurization.gpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.featurization.gpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.featurization.gpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.featurization.gpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.featurization.gpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.featurization.gpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.featurization.gpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.featurization.gpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.featurization.gpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.featurization.gpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.featurization.gpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.featurization.gpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.prediction.cpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.prediction.cpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.prediction.cpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.prediction.cpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.prediction.cpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.prediction.cpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.prediction.cpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.prediction.cpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.prediction.cpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.prediction.cpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.prediction.cpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.prediction.cpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.prediction.cpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.prediction.cpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.prediction.cpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.prediction.cpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.prediction.cpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.prediction.cpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.prediction.cpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.prediction.cpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.prediction.cpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.prediction.cpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.prediction.cpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.prediction.cpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.prediction.cpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.prediction.cpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.prediction.cpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.prediction.cpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.prediction.cpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.prediction.cpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.prediction.cpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.prediction.gpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.prediction.gpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.prediction.gpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.prediction.gpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.prediction.gpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.prediction.gpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.prediction.gpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.prediction.gpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.prediction.gpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.prediction.gpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.prediction.gpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.prediction.gpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.prediction.gpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.prediction.gpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.prediction.gpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.prediction.gpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.prediction.gpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.prediction.gpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.prediction.gpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.prediction.gpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.prediction.gpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.prediction.gpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.prediction.gpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.prediction.gpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.prediction.gpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.prediction.gpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.prediction.gpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.prediction.gpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.prediction.gpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.prediction.gpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.prediction.gpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.training.cpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.training.cpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.training.cpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.training.cpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.training.cpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.training.cpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.training.cpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.training.cpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.training.cpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.training.cpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.training.cpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.training.cpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.training.cpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.training.cpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.training.cpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.training.cpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.training.cpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.training.cpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.training.cpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.training.cpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.training.cpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.training.cpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.training.cpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.training.cpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.training.cpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.training.cpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.training.cpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.training.cpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.training.cpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.training.cpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.training.cpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.jobsTemplates.training.gpu.affinity
Type: core.Affinity [ref]
Affinity defines scheduling constraints for workload
Links:
.spec.jobsTemplates.training.gpu.annotations
Type: object [ref]
Annotations is an unstructured key value map stored with a resource that may be set by external tools to store and retrieve arbitrary metadata. They are not queryable and should be preserved when modifying objects.
Links:
.spec.jobsTemplates.training.gpu.args
Type: array [ref]
Arguments to the entrypoint.
The container image's CMD is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.training.gpu.automountServiceAccountToken
Type: boolean [ref]
AutomountServiceAccountToken indicates whether a service account token should be automatically mounted.
.spec.jobsTemplates.training.gpu.command
Type: array [ref]
Entrypoint array. Not executed within a shell.
The container image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's environment. If a variable
cannot be resolved, the reference in the input string will be unchanged. Double are reduced
to a single $, which allows for escaping the $(VAR_NAME) syntax: i.e. "$$(VAR_NAME)" will
produce the string literal "$(VAR_NAME)". Escaped references will never be expanded, regardless
of whether the variable exists or not. Cannot be updated.
Links:
.spec.jobsTemplates.training.gpu.env
Type: core.EnvVar [ref]
Env keeps the information about environment variables provided to the container
Links:
.spec.jobsTemplates.training.gpu.envFrom
Type: core.EnvFromSource [ref]
EnvFrom keeps the information about environment variable sources provided to the container
Links:
.spec.jobsTemplates.training.gpu.hostIPC
Type: boolean [ref]
HostIPC defines to use the host's ipc namespace.
Default Value: false
.spec.jobsTemplates.training.gpu.hostNetwork
Type: boolean [ref]
HostNetwork requests Host network for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified.
Default Value: false
.spec.jobsTemplates.training.gpu.hostPID
Type: boolean [ref]
HostPID define to use the host's pid namespace.
Default Value: false
.spec.jobsTemplates.training.gpu.image
Type: string [ref]
Image define image details
.spec.jobsTemplates.training.gpu.imagePullPolicy
Type: string [ref]
ImagePullPolicy define Image pull policy
Default Value: IfNotPresent
.spec.jobsTemplates.training.gpu.imagePullSecrets
Type: array [ref]
ImagePullSecrets define Secrets used to pull Image from registry
.spec.jobsTemplates.training.gpu.labels
Type: object [ref]
Map of string keys and values that can be used to organize and categorize (scope and select) objects. May match selectors of replication controllers and services.
Links:
.spec.jobsTemplates.training.gpu.lifecycle
Type: core.Lifecycle [ref]
Lifecycle keeps actions that the management system should take in response to container lifecycle events.
.spec.jobsTemplates.training.gpu.livenessProbe
Type: core.Probe [ref]
LivenessProbe keeps configuration of periodic probe of container liveness. Container will be restarted if the probe fails.
Links:
.spec.jobsTemplates.training.gpu.nodeSelector
Type: object [ref]
NodeSelector is a selector that must be true for the workload to fit on a node.
Links:
.spec.jobsTemplates.training.gpu.ownerReferences
Type: meta.OwnerReference [ref]
List of objects depended by this object. If ALL objects in the list have been deleted, this object will be garbage collected. If this object is managed by a controller, then an entry in this list will point to this controller, with the controller field set to true. There cannot be more than one managing controller.
.spec.jobsTemplates.training.gpu.podSecurityContext
Type: core.PodSecurityContext [ref]
PodSecurityContext holds pod-level security attributes and common container settings.
Links:
.spec.jobsTemplates.training.gpu.ports
Type: []core.ContainerPort [ref]
Ports contains list of ports to expose from the container. Not specifying a port here DOES NOT prevent that port from being exposed. Any port which is listening on the default "0.0.0.0" address inside a container will be accessible from the network.
.spec.jobsTemplates.training.gpu.readinessProbe
Type: core.Probe [ref]
ReadinessProbe keeps configuration of periodic probe of container service readiness. Container will be removed from service endpoints if the probe fails.
Links:
.spec.jobsTemplates.training.gpu.resources
Type: core.ResourceRequirements [ref]
Resources holds resource requests & limits for container
Links:
.spec.jobsTemplates.training.gpu.schedulerName
Type: string [ref]
SchedulerName specifies, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
Default Value: ""
.spec.jobsTemplates.training.gpu.securityContext
Type: core.SecurityContext [ref]
SecurityContext holds container-level security attributes and common container settings.
Links:
.spec.jobsTemplates.training.gpu.serviceAccountName
Type: string [ref]
ServiceAccountName is the name of the ServiceAccount to use to run this pod.
Links:
.spec.jobsTemplates.training.gpu.shareProcessNamespace
Type: boolean [ref]
ShareProcessNamespace defines to share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set.
Default Value: false
.spec.jobsTemplates.training.gpu.startupProbe
Type: core.Probe [ref]
StartupProbe indicates that the Pod has successfully initialized. If specified, no other probes are executed until this completes successfully. If this probe fails, the Pod will be restarted, just as if the livenessProbe failed. This can be used to provide different probe parameters at the beginning of a Pod's lifecycle, when it might take a long time to load data or warm a cache, than during steady-state operation.
Links:
.spec.jobsTemplates.training.gpu.tolerations
Type: []core.Toleration [ref]
Tolerations defines tolerations
Links:
.spec.jobsTemplates.training.gpu.volumeMounts
Type: []core.VolumeMount [ref]
VolumeMounts keeps list of pod volumes to mount into the container's filesystem.
.spec.jobsTemplates.training.gpu.volumes
Type: []core.Volume [ref]
Volumes keeps list of volumes that can be mounted by containers belonging to the pod.
Links:
.spec.jobsTemplates.training.gpu.workingDir
Type: string [ref]
Container's working directory. If not specified, the container runtime's default will be used, which might be configured in the container image.
.spec.metadataService.local.arangoMLFeatureStore
Type: string [ref]
ArangoMLFeatureStoreDatabase define Database name to be used as MetadataService Backend in ArangoMLFeatureStoreDatabase
Default Value: arangomlfeaturestore
.spec.metadataService.local.arangoPipeDatabase
Type: string [ref]
ArangoPipeDatabase define Database name to be used as MetadataService Backend in ArangoPipe
Default Value: arangopipe
.spec.storage.checksum
Type: string [ref]
UID keeps the information about object Checksum
.spec.storage.name
Type: string [ref]
Name of the object
.spec.storage.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.spec.storage.uid
Type: string [ref]
UID keeps the information about object UID
Status
.status.arangoDB.secret.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.arangoDB.secret.name
Type: string [ref]
Name of the object
.status.arangoDB.secret.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.arangoDB.secret.uid
Type: string [ref]
UID keeps the information about object UID
.status.arangoDB.tls.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.arangoDB.tls.name
Type: string [ref]
Name of the object
.status.arangoDB.tls.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.arangoDB.tls.uid
Type: string [ref]
UID keeps the information about object UID
.status.conditions
Type: api.Conditions [ref]
Conditions specific to the entire extension
.status.metadataService.local.arangoMLFeatureStore
Type: string [ref]
ArangoMLFeatureStoreDatabase define Database name to be used as MetadataService Backend
.status.metadataService.local.arangoPipe
Type: string [ref]
ArangoPipeDatabase define Database name to be used as MetadataService Backend
.status.metadataService.secret.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.metadataService.secret.name
Type: string [ref]
Name of the object
.status.metadataService.secret.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.metadataService.secret.uid
Type: string [ref]
UID keeps the information about object UID
.status.reconciliation.service.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.reconciliation.service.name
Type: string [ref]
Name of the object
.status.reconciliation.service.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.reconciliation.service.uid
Type: string [ref]
UID keeps the information about object UID
.status.reconciliation.statefulSet.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.reconciliation.statefulSet.name
Type: string [ref]
Name of the object
.status.reconciliation.statefulSet.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.reconciliation.statefulSet.uid
Type: string [ref]
UID keeps the information about object UID
.status.serviceAccount.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.serviceAccount.cluster.binding.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.serviceAccount.cluster.binding.name
Type: string [ref]
Name of the object
.status.serviceAccount.cluster.binding.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.serviceAccount.cluster.binding.uid
Type: string [ref]
UID keeps the information about object UID
.status.serviceAccount.cluster.role.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.serviceAccount.cluster.role.name
Type: string [ref]
Name of the object
.status.serviceAccount.cluster.role.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.serviceAccount.cluster.role.uid
Type: string [ref]
UID keeps the information about object UID
.status.serviceAccount.name
Type: string [ref]
Name of the object
.status.serviceAccount.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.serviceAccount.namespaced.binding.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.serviceAccount.namespaced.binding.name
Type: string [ref]
Name of the object
.status.serviceAccount.namespaced.binding.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.serviceAccount.namespaced.binding.uid
Type: string [ref]
UID keeps the information about object UID
.status.serviceAccount.namespaced.role.checksum
Type: string [ref]
UID keeps the information about object Checksum
.status.serviceAccount.namespaced.role.name
Type: string [ref]
Name of the object
.status.serviceAccount.namespaced.role.namespace
Type: string [ref]
Namespace of the object. Should default to the namespace of the parent object
.status.serviceAccount.namespaced.role.uid
Type: string [ref]
UID keeps the information about object UID
.status.serviceAccount.uid
Type: string [ref]
UID keeps the information about object UID