For AI agents: the complete documentation index is available at https://docs.us.ovhcloud.com/en/llms.txt, the full documentation bundle is available at https://docs.us.ovhcloud.com/en/llms-full.txt, and this page is available as Markdown at https://docs.us.ovhcloud.com/en/bare-metal-cloud/dedicated-servers/hardware-diagnostics-dedicated-server.md.
  • English

Hardware diagnostics for a dedicated server

View as Markdown

Learn how to diagnose hardware issues on your server.

At some point during the life of your server, you may encounter a fault due to a hardware issue. When the server is booted into the OVHcloud rescue mode, you have several diagnostic tools available to help identify faulty hardware components.

 

Requirements

 

Instructions

This guide details the tests you need to carry out to diagnose:

  • Processors
  • Network connection
  • Memory
  • Disk health
  • Disk partitions

 

Processors

The processor tests check the working order of your server's processor and need about 30 minutes to run successfully. If the server crashes during this test, it could indicate issues with the CPU, RAM, or motherboard.

To check the CPU:

WRKR=$(grep -c "^processor" /proc/cpuinfo)
stress-ng --metrics-brief --timeout 60s --cpu $WRKR --io $WRKR --aggressive --ignite-cpu --maximize --pathological

To check the RAM:

stress-ng --metrics-brief --timeout 60s --brk 0 --stack 0 --bigheap 0

 

Network Connection

The network connection test checks your internal and external bandwidth. This data is provided for information purposes only and is not a performance test.

ping -c 10 proof.ovh.us
for file in 1Mb 10Mb 100Mb 1Gb ; do time curl -4f https://proof.ovh.us/files/${file}.dat -o /dev/null; done

 

Memory

The memory test checks the integrity of your server's RAM modules. If the server crashes during this test, then it means that one or more of your RAM modules is faulty.

Warning

NOTE: This test can take a long time to complete.

RAM="$(awk -vOFMT=%.0f '$1 == "MemAvailable:" {print $2/1024 - 1024}' /proc/meminfo)"
memtester ${RAM}M 1

 

Disk Health

You can use Smartmontools to check the status of your disks by reading their SMART data. For example, to display all the details of the disk labeled nvme1n1, enter:

smartctl -a /dev/nvme1n1

To learn more about the output of this command and how to interpret it, consult the official Smartmontools documentation.

 

Disk Partitions

The partitions test is comprised of a disk access test and a file system check. The disk access test checks if the system can communicate with your server's hard drives. The file system check uses the fsck -fy command to check the entire file system.

To check the file system, run the following command:

fsck-fy

For a read test, run the command below. Replace sd(x) with your own values.

hdparm -t /dev/sd(x)

 

Go further

Browse all the Dedicated Servers guides.

Browse the rest of the OVHcloud US documentation for guides on our other products and services.

Learn more about Dedicated Servers on the OVHcloud US website.