dcgm-exporter helm

No introduction found. Create it?

Install

Install with:

helm repo add dcgm-exporter oci://ghcr.io/bjw-s-labs/charts/
helm install dcgm-exporter dcgm-exporter/dcgm-exporter -f values.yaml

Examples

See examples from other people.

Top Repositories (3 out of 8)

NameRepoStarsVersionTimestamp
dcgm-exportermirceanton/home-ops1224.8.49 hours ago
dcgm-exporterchr1sd/home-ops1935.2.12 days ago
dcgm-exporterjoryirving/home-ops2275.2.14 months ago

Resources

128Mi256Mi
512Mi768Mi1024Mi

Values

See the most popular values for this chart:

KeyTypes
resources.limits.memory (5)
768Mi
string
resources.limits."nvidia.com/gpu" (4)
1
number
resources.requests.memory (5)
128Mi
string
resources.requests.cpu (2)
50m
string
affinity.nodeAffinity.requiredDuringSchedulingIgnoredDuringExecution.nodeSelectorTerms[].matchExpressions[].key (4)
extensions.talos.dev/nvidia-open-gpu-kernel-modules-production
string
affinity.nodeAffinity.requiredDuringSchedulingIgnoredDuringExecution.nodeSelectorTerms[].matchExpressions[].operator (4)
Exists
string
serviceMonitor.honorLabels (4)
true
boolean
serviceMonitor.interval (3)
15s
string
serviceMonitor.enabled (2)
true
boolean
serviceMonitor.scrapeTimeout (2)
10s
string
serviceMonitor.app.endpoints[].honorLabels (1)
true
boolean
serviceMonitor.app.endpoints[].interval (1)
30s
string
serviceMonitor.app.endpoints[].port (1)
metrics
string
serviceMonitor.app.endpoints[].relabelings[].action (1)
replace
string
serviceMonitor.app.endpoints[].relabelings[].sourceLabels[] (1)
- __meta_kubernetes_pod_node_name
string
serviceMonitor.app.endpoints[].relabelings[].targetLabel (1)
instance
string
serviceMonitor.app.endpoints[].scrapeTimeout (1)
10s
string
serviceMonitor.labels.prometheus (1)
kube-prometheus-stack
string
serviceMonitor.relabelings[].action (1)
replace
string
serviceMonitor.relabelings[].regex (1)
(.*)
string
serviceMonitor.relabelings[].replacement (1)
$1.internal
string
serviceMonitor.relabelings[].sourceLabels[] (1)
- __meta_kubernetes_endpoint_node_name
string
serviceMonitor.relabelings[].targetLabel (1)
instance
string
controllers.dcgm-exporter.annotations."reloader.stakater.com/auto" (3)
true
string
controllers.dcgm-exporter.containers.app.args[] (3)
- -f
- /etc/dcgm-exporter/default-counters.csv
string
controllers.dcgm-exporter.containers.app.env.DCGM_EXPORTER_KUBERNETES (3)
true
string
controllers.dcgm-exporter.containers.app.env.DCGM_EXPORTER_LISTEN (3)
:9400
string
controllers.dcgm-exporter.containers.app.env.NODE_NAME.valueFrom.fieldRef.fieldPath (3)
spec.nodeName
string
controllers.dcgm-exporter.containers.app.env.KUBERNETES_ENABLE_DRA (1)
true
string
controllers.dcgm-exporter.containers.app.env.NVIDIA_DRIVER_CAPABILITIES (1)
utility
string
controllers.dcgm-exporter.containers.app.env.NVIDIA_VISIBLE_DEVICES (1)
all
string
controllers.dcgm-exporter.containers.app.image.repository (3)
nvcr.io/nvidia/k8s/dcgm-exporter
string
controllers.dcgm-exporter.containers.app.image.tag (3)
4.5.3-4.8.2-distroless@sha256:e9030b4fca0c8f110032f3b151b030e7e9db56604b407cf79631e1a3218b35e0
string
controllers.dcgm-exporter.containers.app.probes.liveness.custom (3)
true
boolean
controllers.dcgm-exporter.containers.app.probes.liveness.enabled (3)
true
boolean
controllers.dcgm-exporter.containers.app.probes.liveness.spec.httpGet.path (3)
/health
string
controllers.dcgm-exporter.containers.app.probes.liveness.spec.httpGet.port (3)
9400
number
controllers.dcgm-exporter.containers.app.probes.liveness.spec.initialDelaySeconds (3)
45
number
controllers.dcgm-exporter.containers.app.probes.liveness.spec.periodSeconds (3)
15
number
controllers.dcgm-exporter.containers.app.probes.liveness.spec.timeoutSeconds (3)
5
number
controllers.dcgm-exporter.containers.app.probes.readiness.custom (3)
true
boolean
controllers.dcgm-exporter.containers.app.probes.readiness.enabled (3)
true
boolean
controllers.dcgm-exporter.containers.app.probes.readiness.spec.httpGet.path (3)
/health
string
controllers.dcgm-exporter.containers.app.probes.readiness.spec.httpGet.port (3)
9400
number
controllers.dcgm-exporter.containers.app.probes.readiness.spec.initialDelaySeconds (3)
45
number
controllers.dcgm-exporter.containers.app.probes.readiness.spec.periodSeconds (3)
15
number
controllers.dcgm-exporter.containers.app.probes.readiness.spec.timeoutSeconds (3)
5
number
controllers.dcgm-exporter.containers.app.resources.limits.memory (3)
512Mi
string
controllers.dcgm-exporter.containers.app.resources.limits.cpu (2)
200m
string
controllers.dcgm-exporter.containers.app.resources.requests.cpu (3)
100m
string
controllers.dcgm-exporter.containers.app.resources.requests.memory (3)
128Mi
string
controllers.dcgm-exporter.containers.app.securityContext.allowPrivilegeEscalation (3)
false
boolean
controllers.dcgm-exporter.containers.app.securityContext.capabilities.drop[] (3)
- ALL
string
controllers.dcgm-exporter.containers.app.securityContext.runAsGroup (3)
0
number
controllers.dcgm-exporter.containers.app.securityContext.runAsNonRoot (3)
false
boolean
controllers.dcgm-exporter.containers.app.securityContext.runAsUser (3)
0
number
controllers.dcgm-exporter.type (3)
daemonset
string
defaultPodOptions.runtimeClassName (3)
nvidia
string
defaultPodOptions.tolerations[].effect (3)
NoSchedule
string
defaultPodOptions.tolerations[].key (3)
llm-workload
string
defaultPodOptions.tolerations[].operator (3)
Exists
string
defaultPodOptions.automountServiceAccountToken (2)
true
boolean
defaultPodOptions.nodeSelector."topology.kubernetes.io" (2)
dgpu
string
defaultPodOptions.nodeSelector."nvidia.com/gpu.present" (1)
true
string
persistence.exporter-metrics.globalMounts[].path (3)
/etc/dcgm-exporter/default-counters.csv
string
persistence.exporter-metrics.globalMounts[].readOnly (3)
true
boolean
persistence.exporter-metrics.globalMounts[].subPath (3)
default-counters.csv
string
persistence.exporter-metrics.name (3)
dcgm-exporter-metrics
string
persistence.exporter-metrics.type (3)
configMap
string
persistence.pod-gpu-resources.globalMounts[].path (2)
/var/lib/kubelet/pod-resources
string
persistence.pod-gpu-resources.globalMounts[].readOnly (2)
true
boolean
persistence.pod-gpu-resources.hostPath (2)
/var/lib/kubelet/pod-resources
string
persistence.pod-gpu-resources.type (2)
hostPath
string
persistence.pod-resources.globalMounts[].path (1)
/var/lib/kubelet/pod-resources
string
persistence.pod-resources.globalMounts[].readOnly (1)
true
boolean
persistence.pod-resources.hostPath (1)
/var/lib/kubelet/pod-resources
string
persistence.pod-resources.type (1)
hostPath
string
runtimeClassName (3)
nvidia
string
service.app.ports.metrics.port (3)
9400
number
service.app.ports.metrics.enabled (2)
true
boolean
extraEnv[].name (2)
NVIDIA_DRIVER_CAPABILITIES
NVIDIA_VISIBLE_DEVICES
string
extraEnv[].value (2)
all
all
string
fullnameOverride (2)
dcgm-exporter
string
tolerations[].effect (2)
NoSchedule
string
tolerations[].key (2)
nvidia.com/gpu
string
tolerations[].operator (2)
Exists
string
image.repository (1)
mirror.gcr.io/nvidia/dcgm-exporter
string
image.tag (1)
4.5.1-4.8.0-distroless
string
kubernetesDRA.enabled (1)
true
boolean
rbac.bindings.dcgm-exporter.roleRef.identifier (1)
dcgm-exporter
string
rbac.bindings.dcgm-exporter.subjects[].identifier (1)
dcgm-exporter
string
rbac.bindings.dcgm-exporter.type (1)
ClusterRoleBinding
string
rbac.roles.dcgm-exporter.rules[].apiGroups[] (1)
-
- resource.k8s.io
string
rbac.roles.dcgm-exporter.rules[].resources[] (1)
- pods
- resourceslices
string
rbac.roles.dcgm-exporter.rules[].verbs[] (1)
- get
- list
- watch
string
rbac.roles.dcgm-exporter.type (1)
ClusterRole
string
securityContext.allowPrivilegeEscalation (1)
true
boolean
securityContext.capabilities.add[] (1)
- SYS_ADMIN
string
securityContext.privileged (1)
true
boolean