NVIDIA/gpu-monitoring-tools

Issues

• dcgm-exporter falied so start on GKE cluster (v1.16.11-gke.5)

  #96 opened 4 years ago by Dimss
  12

• dcgm-exporter missing many metrics after upgrade

  #143 opened 4 years ago by huww98
  9

• Error watching fields: Profiling is not supported for this group of GPUs or GPU

  #119 opened 4 years ago by motionlife
  10

• Container, namespace and pod informations on metrics

  #138 opened 4 years ago by josericardomcastro
  9

• dcgm-exporter high cpu usage

  #133 opened 4 years ago by ysshaoxiao
  6

• why require k8s.io/kubernetes project directly, is not recommended

  #123 opened 4 years ago by utobe67
  12

• Get GPU busid failed by changing default metric csv

  #118 opened 4 years ago by pokerfaceSad
  1

• No labels with GPU-Card Name in dcgm-exporter

  #91 opened 4 years ago by vizdrag
  1

• Pods with dcgm-exporter fail to start

  #120 opened 4 years ago by timClicks
  12

• how prometheus get dcgm-exporter metrics?

  #106 opened 4 years ago by Leteong
  12

• dcgm-exporter doesn't start on Docker

  #134 opened 4 years ago by gurapomu
  3

• Grafana: GPU power total gauge, sum not useful

  #109 opened 4 years ago by mjpieters
  5

• Custom metrics issue

  #122 opened 4 years ago by PaulYuanJ
  4

• Prometheus Exporters Hub by this repository! Thanks! :)

  #144 opened 4 years ago by ralfyang
  0

• dcgm-exporter: DCP metrics not enabled

  #140 opened 4 years ago by jelmd
  2

• Erro start dcgm-exporter pod - module of DCGM that is not currently loaded

  #141 opened 4 years ago by josericardomcastro
  1

• dcgm-exporter crashes after MIG reconfiguration

  #142 opened 4 years ago by kpouget
  1

• nvmlShutdown dlcloses all handles every time

  #139 opened 4 years ago by robertdavidsmith
  0

• Allow helm chart to customize kubelet path

  #137 opened 4 years ago by vdebergue
  0

• Error checking GPU health: API version mismatch

  #135 opened 4 years ago by noliaoliao
  0

• K8s Pod/namespace information in exported fields

  #129 opened 4 years ago by geoberle
  4

• Failed to install gpu-helm-charts/dcgm-exporter

  #132 opened 4 years ago by jasperzhong
  1

• Consider adding a 'pod' labels，which aggregate data at prometheus？

  #131 opened 4 years ago by qingwei8
  0

• whether A100 mig is supported

  #127 opened 4 years ago by zhcf
  1

• FR: Expose option to use JSON logs

  #128 opened 4 years ago by etherandrius
  0

• Consider adding a 'nodename' label

  #110 opened 4 years ago by mjpieters
  1

• Exposed metrics don't follow Prometheus spec

  #126 opened 4 years ago by etherandrius
  0

• Sudden error message about /run/prometheus/dcgm.prom

  #125 opened 4 years ago by kubernetian
  0

• Error getting process info: Setting not configured

  #116 opened 4 years ago by CermakM
  0

• Unable to start according to the instructions

  #112 opened 4 years ago by zkbutt
  1

• Examples not available in the latest version of DCGM

  #115 opened 4 years ago by Tabrizian
  0

• DCGM Python3 bindings

  #114 opened 4 years ago by Tabrizian
  0

• dcgm-exporter POD CrashLoopBackOff or Error

  #111 opened 4 years ago by Leteong
  0

• Failed to initialize NVML

  #90 opened 4 years ago by guleng
  1

• Broken link in README

  #101 opened 4 years ago by kshcherban
  2

• Wrong value in DCGM_FI_DEV_MEMORY_TEMP for NVIDIA Tesla-T4

  #97 opened 4 years ago by maxbischoff
  1

• dcgm-exporter POD CrashLoopBackOff or Error

  #105 opened 4 years ago by Leteong
  0

• nvml.h | Request for Fan Speed RPM (not percent) | NV_CTRL_THERMAL_COOLER_SPEED

  #107 opened 4 years ago by berglh
  0

• all my data is '0'

  #104 opened 4 years ago by darkamumu
  0

• Bare Metal | /run/prometheus/dcgm.prom Not Present

  #103 opened 4 years ago by atulyadavtech
  0

• dcgm-exporter `Error getting device information: API version mismatch`

  #102 opened 4 years ago by notjames
  2

• Add label for kubernetes node to enable data aggregation

  #99 opened 4 years ago by MartinForReal
  0

• Make the helm chart available via hub.helm.sh

  #98 opened 4 years ago by patrungel
  1

• dcgm-exporter POD cannot be running

  #94 opened 4 years ago by nakkoh
  0

• Option to pass hostname/ip along with port

  #85 opened 4 years ago by bbelgodere
  3

• method is not support test suit

  #88 opened 4 years ago by cuisongliu
  0

• TLS Support

  #80 opened 4 years ago by RenaudWasTaken
  1

• Fan status and card count requirements

  #84 opened 4 years ago by guleng
  2

• unknown flag

  #81 opened 4 years ago by guleng
  3

• dcgm exporter produces 404 page not found

  #78 opened 4 years ago by bbelgodere
  7