Common Refurbished Server Problems and How to Fix Them

Server Repairs

Most problems with a refurbished server show up in the first days of use, and nearly all of them are easy to diagnose because enterprise servers log their own faults. This guide covers the issues we see most often on refurbished Dell PowerEdge and HPE ProLiant servers, how to find the cause, and how to fix each one.

Quick answer: the most common issues are a failed or failing drive, a degraded RAID array, memory errors, a failed power supply or fan, overheating, and setup problems such as the operating system not seeing the disks. Check the management log first (iDRAC on Dell, iLO on HPE). It usually names the exact part that needs attention.

Start with the server’s own logs

Before you open the case, read what the server has already recorded. Dell PowerEdge servers keep a System Event Log and a Lifecycle Log in iDRAC. HPE ProLiant servers keep the Integrated Management Log (IML) in iLO. Both list failed drives, memory errors, power supply faults, fan failures and temperature warnings with a time stamp and the component location.

Pro tip: set up email alerts in iDRAC or iLO on day one. You will hear about a failing drive or power supply before it causes downtime.

Common refurbished server issues and fixes

SymptomLikely causeWhat to do
Amber light on a drive, or RAID shows “degraded”Failed or failing hard driveReplace the drive with a compatible model and let the array rebuild. Check the rebuild completes in the RAID controller.
Memory error in the log, or the server shows less RAM than installedFaulty or unseated DIMM, or incorrect slot populationReseat the module, check the slot population rules for your model, and replace a DIMM that keeps logging errors.
Power supply fault, or the server says redundancy is lostFailed PSU, or one PSU not connectedCheck both power cords and the outlet, then replace the failed PSU. Hot-swap PSUs can be changed while the server runs.
Fans run at full speed all the timeFailed fan, missing blanking panel, high temperature, or a non-standard PCIe cardCheck the log for fan or temperature errors, fit all blanks and covers, and check whether a third-party card is causing a higher fan profile.
Server shuts down or throttlesOverheatingImprove airflow, clean intakes and keep the room cool. See our guide to keeping a server cool.
Operating system installer cannot see the disksMissing RAID controller driver, or no virtual disk createdCreate a virtual disk in the RAID controller first, then load the controller driver or use Dell Lifecycle Controller or HPE Intelligent Provisioning to deploy the OS.
Date, time or BIOS settings reset after power lossFlat CMOS batteryReplace the coin cell battery and set the BIOS again.
Server will not boot from the diskBoot mode mismatch (UEFI vs legacy BIOS) or wrong boot orderMatch the boot mode to how the OS was installed and set the boot order.
server maintenance

How to troubleshoot step by step

  1. Read the log. iDRAC or iLO usually tells you exactly what failed and where.
  2. Check the obvious. Power cords, network cables, seated drives and modules, and that covers and blanks are fitted.
  3. Check firmware. Mismatched BIOS, controller and drive firmware can cause odd behaviour. Read our guide to firmware updates.
  4. Run the built-in diagnostics. Dell servers include ePSA or SupportAssist diagnostics in the Lifecycle Controller, and HPE servers include diagnostics in Intelligent Provisioning.
  5. Swap one part at a time. Change a single DIMM, drive or PSU and test again, so you know which part fixed it.
Watch out: do not pull a drive from a degraded RAID array to test it unless you are sure which drive has failed. Removing the wrong drive can take the whole array offline. Use the amber fault light and the controller to identify it.

How to prevent problems

  • Run drives in RAID and keep a spare drive on hand. Our RAID guide explains the options.
  • Use redundant power supplies on a UPS, especially during load shedding.
  • Keep the room cool and dust-free.
  • Update firmware during planned maintenance windows.
  • Monitor the server so you see problems early. Compare free tools in our server monitoring guide.
  • Back up regularly and test that restores work.
Good to know: a properly refurbished server should arrive tested, with its logs checked and worn parts replaced. Read how refurbished servers are tested, and see what is covered under our refurbished server warranty.

When to call for help

Contact your supplier if a part fails within the warranty period, if the same error keeps returning after you replace a part, or if the server will not complete its power-on self test. Have the log entry, the server model and the service tag or serial number ready. For general upkeep, follow our refurbished server maintenance guide.

Frequently asked questions

What is the most common problem with refurbished servers?

Hard drive failures are the most common hardware issue on any server, new or refurbished, because drives have moving parts. Running drives in RAID means one drive can fail without data loss.

How do I check a server for hardware faults?

Open the management interface, iDRAC on Dell or iLO on HPE, and read the event logs. They record failed drives, memory errors, power supply faults and temperature warnings.

Why are my server fans so loud?

Fans run at full speed briefly at start-up, which is normal. If they stay loud, check the log for a failed fan, high temperatures, missing blanking panels or a third-party PCIe card that triggers a higher fan profile.

Why can my operating system installer not see the server disks?

Usually no virtual disk has been created in the RAID controller, or the installer needs the controller driver. Create the virtual disk first, then use Dell Lifecycle Controller or HPE Intelligent Provisioning, or load the driver manually.

Need a replacement drive, memory or power supply? Browse our server parts, server hard drives and server memory, or contact our team.

Server Site