docs: add GPU Operator production troubleshooting runbook#2198
Open
framsouza wants to merge 2 commits intoNVIDIA:mainfrom
Open
docs: add GPU Operator production troubleshooting runbook#2198framsouza wants to merge 2 commits intoNVIDIA:mainfrom
framsouza wants to merge 2 commits intoNVIDIA:mainfrom
Conversation
framsouza
requested review from
cdesiniotis,
karthikvetrivel,
rahulait,
rajathagasthya,
shivamerla and
tariq1890
as code owners
framsouza
force-pushed
the
gpu-troubleshooting-runbook
branch
from
2075c4f to
6986bcc
Compare
The guide covers scenarios such as: - GPUs not detected on nodes - GPU workloads stuck in Pending - driver daemonset failures - device plugin initialization issues - missing DCGM exporter metrics - MIG configuration problems It also includes example debugging commands and sample outputs to help operators quickly recognize common failure patterns. The content is based on past operational troubleshooting experiences running GPU workloads on Kubernetes clusters using the GPU Operator. Signed-off-by: framsouza <fram.souza14@gmail.com>
Signed-off-by: framsouza <fram.souza14@gmail.com>
framsouza
force-pushed
the
gpu-troubleshooting-runbook
branch
from
b34e577 to
4832201
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds a production troubleshooting runbook for common operational issues I encountered when running the NVIDIA GPU Operator in Kubernetes clusters.
The runbook includes step-by-step debugging guidance for situations such as:
PendingEach section includes example commands and mock outputs to help operators quickly identify common failure patterns.
GPU clusters can be challenging to troubleshoot when components such as drivers, container runtime integration, or the device plugin fail to initialize correctly.
This runbook aims to provide a structured debugging flow that platform engineers and SREs can follow during incidents (also if you're curious about the operator).
The content is based on past troubleshooting experiences operating GPU workloads on Kubernetes clusters using the GPU Operator.