Software⏱️ 4 min read📅 2026-06-11

How to Fix: "Graphic card error(nvidia-smi prints "ERR!" on FAN and Usage)" and processes are not killed and gpu not being reset

Technical issue with GPU error and processes not being killed.

Quick Answer: Try killing the processes using "sudo kill -9 pid" again, or use a process manager like systemd to manage GPU usage.

📋 Table of Contents

  1. 🛑 Root Causes of the Error
  2. ✅ Best Solutions to Fix It
  3. ✨ Wrapping Up

The issue described is a common problem encountered by users who are using NVIDIA GPUs in Ubuntu servers. The error message 'ERR!' printed by 'nvidia-smi' indicates an issue with the GPU fan and usage, preventing the GPU from being reset or killed. This can be frustrating for users who need to run multiple codes on one GPU, as it can lead to resource conflicts and performance issues.

In this guide, we will walk you through the steps to troubleshoot and resolve this issue. We will cover the root causes of the problem, provide two primary fix methods, and offer additional tips and considerations for users who are new to using GPUs.

🛑 Root Causes of the Error

  • The first main reason why this error happens is that the NVIDIA GPU drivers or system configuration may not be properly set up. This can cause conflicts between different processes running on the same GPU, leading to resource allocation issues and the 'ERR!' error message.
  • Another alternative reason for this issue could be due to incorrect usage of the 'gpustat' command or other tools that monitor GPU usage. Running multiple programs using one GPU without properly managing resources can lead to conflicts and performance issues.

✅ Best Solutions to Fix It

Killing Zombie Processes

  1. Step 1: To start, identify any Zombie processes running on the system using the 'ps -eo pid,ppid,lstart,cmd' command. These processes are no longer responding to signals and can cause issues with GPU usage.
  2. Step 2: Once you have identified Zombie processes, use the 'kill' command to terminate them. Be cautious when using the '-9' flag, as it will forcefully terminate the process without giving it a chance to clean up resources.
  3. Step 3: After killing the Zombie processes, try running 'nvidia-smi --gpu-reset -i 0' again to reset the GPU index 0.

Killing Processes Running on the GPU

  1. Step 1: If the above method does not work, you can try killing the processes that are running on the GPU using 'sudo kill -9 '. However, be aware that this may not always work, especially if the process is a background job or has elevated privileges.
  2. Step 2: To avoid this issue in the future, consider using tools like 'nvidia-smi' to monitor GPU usage and terminate any unnecessary processes. You can also use 'gpustat' to list all running processes on the GPU and identify which ones are causing issues.

✨ Wrapping Up

In conclusion, the 'ERR!' error message printed by 'nvidia-smi' is a common issue that can be resolved by killing Zombie processes or terminating processes running on the GPU. By following the steps outlined in this guide, you should be able to resolve the issue and get your system up and running smoothly.

Did this fix your problem?

If not, try searching for specific error codes.

🔍 Search Error Database

❓ Frequently Asked Questions

🛠️ Related Fixes

How to Fix: Pc crashes shortly after launching game (rainbow

Fix Pc crashes shortly after launching game (rainbow six siege). Compl

How to Fix: Installing an APK on a locked down phone

Installing an APK on a locked down phone: Try using a rooted device, e

How to Fix: New PC build- no signal and no clue

Fix New PC build- no signal and no clue. Complete troubleshooting guid