Issues

• [Feature] Optional PrometheusRule in the Helm chart for lost GPU telemetry

  NVIDIA/dcgm-exporter#766 · Filter by author Sajish85 opened on Nov 1, 2026

• [ALERT] Access to NVIDIA's Self-Hosted Runners is Expiring

  nvidia-runners
  NVIDIA/dcgm-exporter#765 · Filter by author nvidia-runner-mgmt opened on Oct 28, 2026

• dcgm-exporter exits on MIG GPUs when NVML lacks nvmlDeviceGetGpuInstanceProfileInfoByIdV

  bug
  NVIDIA/dcgm-exporter#757 · Filter by author kkohtaka opened on Oct 13, 2026
  1

• Pod label normalization silently drops colliding Kubernetes keys

  NVIDIA/dcgm-exporter#754 · Filter by author git-jxj opened on Oct 12, 2026

• GPU NVLink link index is renumbered from 1, while NvSwitch links keep DCGM's index

  NVIDIA/dcgm-exporter#753 · Filter by author v0ropaev opened on Oct 10, 2026
  1

• XID errors never exported on MIG GPUs (DCGM stores them only on compute instances)

  NVIDIA/dcgm-exporter#750 · Filter by author erikkressner opened on Oct 2, 2026
  1

• DCGM 4.7 Support

  NVIDIA/dcgm-exporter#743 · Filter by author daveoy opened on Sep 16, 2026
  1

• Multiple host engine support for dcgm-exporter

  question
  NVIDIA/dcgm-exporter#742 · Filter by author Sukkaito opened on Sep 16, 2026

• A100 enable mig, some metrics missing(DCGM_FI_PROF_PCIE_RX/TX_BYTES,DCGM_FI_PROF_NVLINK_RX/TX_BYTES)

  bug
  NVIDIA/dcgm-exporter#726 · Filter by author magiceses opened on Aug 31, 2026
  2

• Add optional cgroup-direct process attribution for non-MIG shared GPUs

  enhancement
  NVIDIA/dcgm-exporter#725 · Filter by author SMS-Derfflinger opened on Aug 26, 2026
  1

• DCGM_FI_PROF_PCIE_TX_BYTES and DCGM_FI_PROF_PCIE_RX_BYTES counters are always zero

  bug
  NVIDIA/dcgm-exporter#721 · Filter by author joseluisgonzalezca opened on Aug 11, 2026
  3

• Missing DRA metrics in case of dynamically created MIG devices

  NVIDIA/dcgm-exporter#714 · Filter by author guptaNswati opened on Jul 23, 2026
  3

• FI_PROF_* metrics not exported (profiling module is not currently loaded)

  question
  NVIDIA/dcgm-exporter#707 · Filter by author tensorriccii opened on Jul 2, 2026
  1

• dcgm-exporter 4.4.2 memory leak causing periodic OOM kills

  bug
  NVIDIA/dcgm-exporter#702 · Filter by author inyohh opened on Jun 29, 2026
  3

• Visibility of per-slice NVENC/NVDEC metrics under MIG profiles

  enhancement
  NVIDIA/dcgm-exporter#675 · Filter by author tulipaws opened on Jun 11, 2026

• DCGM Exporter doesn't work with fractional vGPU shapes (nvidia-rtx-pro-6000)

  bug
  NVIDIA/dcgm-exporter#661 · Filter by author JeffLuoo opened on May 19, 2026

• 4.5.2-4.8.1 fails to start on node with A30 (MIG enabled) and RTX2000

  bug
  NVIDIA/dcgm-exporter#657 · Filter by author gfrankliu opened on May 9, 2026
  5

• Support DCGM_FT_BINARY field types as Prometheus metrics

  enhancement
  NVIDIA/dcgm-exporter#645 · Filter by author erdinc-yukarikir opened on Mar 25, 2026

• Where is a list of metrics with their meanings?

  question
  NVIDIA/dcgm-exporter#643 · Filter by author smktpd opened on Mar 10, 2026
  3

• Show Hostname in the Grafana Dashboard graph labels to prevent confusion on the GPU number

  enhancement
  NVIDIA/dcgm-exporter#630 · Filter by author pamudithaA opened on Feb 12, 2026

• NVML library init failure on GB200/GB300 ARM64 containers

  bug
  NVIDIA/dcgm-exporter#621 · Filter by author vkvenkat opened on Jan 26, 2026
  1

• Helm chart fails on EKS with Bottlerocket

  bug
  NVIDIA/dcgm-exporter#613 · Filter by author tomershaiman opened on Jan 14, 2026
  1

• 4.3.1-4.4.1 Not collecting DCP metrics in 470.256.02 driver

  bug
  NVIDIA/dcgm-exporter#608 · Filter by author LastNight1997 opened on Jan 7, 2026

• Stop or Restart of nvidia-dcgm-exporter.service hangs

  bug
  NVIDIA/dcgm-exporter#606 · Filter by author fabienarchambault opened on Dec 23, 2025

• Dcgm-exporter not working

  bug
  NVIDIA/dcgm-exporter#603 · Filter by author gggh900 opened on Dec 16, 2025