Importing HDDs from a failed server

Last Updated : Jan 26, 2026 |

About this task

If the user is attempting an HDD import from the failed server’s HDDs follow these steps below. If a RAID configuration is required and an import is not possible, continue to the RAID configuration chapter.

This procedure applies to both the R640 and R660xs. Import allows for a faster recovery of a server. The assumption is that the HDDs from the failed server are good. If the HDDs are good, then an import is possible.

Note:

Screenshots used in this procedure are for illustration purposes only.

Procedure

  1. With power cords still removed on both systems, remove all the HDDs from the FRU server and set them aside. All HDDs from the FRU server must be removed before proceeding with the import process. The HDDs that have been removed from the FRU server will not be needed assuming all HDDs from the failed server are still operational.
  2. Remove the HDDs one at a time from the failed server and insert them into the corresponding slot on the FRU server.
  3. Once all HDDs have been moved to the FRU server, confirm that the new FRU server is the identical hardware construct as the failed server that it is replacing.
  4. Once FRU server hardware configuration is confirmed to match the failed server, install back into rack position.
  5. Connect all other media connections to the server (NIC cables to same NIC ports of the previous failed server).
  6. Connect video monitor, USB keyboard, and mouse to server.
  7. Insert power cords into the replacement server.
  8. Press the power button located at the front right corner of the server and view the monitor screen. This is referred to as the server console.

    If the RAID import is successful, the metadata resident on the HDDs should configure the replacement RAID controller to match the previous RAID controller that the HDDs were connected to. The server should progress through the boot up process and load the operating system.

  9. When the server powers up the hardware boot screen displays (below). It does not indicate that the import was successful. Continue watching the monitor as the server attempts to boot from the newly inserted HDDs.
    Note:

    The BIOS version varies depending on the server hardware type (for example, R660xs vs. R640).





  10. The following screen appears as the server attempts to boot from Red Hat Enterprise Linux for the first time. It is possible that the message Boot Failed may be displayed. The server will continue to attempt to boot. This message may appear again. No manual intervention is required. Allow the server to continue to boot. During this period, the system may pause for several seconds. This behavior is expected, as the server is in the process of importing the RAID configuration and associated data from the foreign hard drives (HDDs).




  11. Once the import process is complete, seeing the following screen is a good indication that the RAID configuration, along with all its data, was successfully imported. At this point, the RAID controller has assumed the RAID configuration from the failed server. Continue to allow the system to boot from its default highlighted OS selection.




  12. If the screen shown below appears, it indicates that KVM on Red Hat Enterprise Linux 8.10 has successfully booted on the server. This indicates that the RAID import was successful and that the data on the hard drives (HDDs) remains intact. At this stage, the RAID controller has correctly adopted the RAID configuration from the failed server.

    The next step is for the user to log in to the Cockpit host interface and verify the following:

    1. Host networking configuration was not altered during the import process.

    2. Validate virtual machines are present. Power on VMs as required if the AutoStart feature has not been set.

    Move on to step 13 below. If the screen below does not appear, proceed to Step 14 to retry the import procedure.





  13. Verify Host configuration was not altered during import of the HDDs.
    1. Log in to cockpit using the Host services port connection and authenticate with existing custadm credentials.
    2. Navigate to the Networking page and validate every expected interface is being displayed, such as bridge, bond, and VLAN tags.




    3. Navigate to the Virtual machines page, and validate that you can see all expected virtual machines. Power on virtual machines on demand as required if AutoStart feature has not been set.




  14. After Host network configuration and virtual machine verification, the RAID controller battery that came in the FRU server should have its health checked as outlined below in steps a – c. Refer to the RAID Battery chapter for additional information.
    1. Verify Raid battery is good by executing step: Checking the RAID battery health using system setup menu. If RAID battery health is good, then import task is complete.
    2. If the RAID battery requires replacement move the RAID battery from failed server to the replacement FRU server and recheck battery health. If battery health is good, import task is complete.
    3. If the RAID battery requires replacement, then a new RAID battery will need to be procured. Go to section RAID Battery ordering information.
  15. If the screen below appears then ASP 130 R6.0.x (KVM on RHEL 8.10) did not load, a RAID configuration setting should be verified. Select F2 to enter System Setup.




RAID Import Verification Setting

The HDD import may have failed because the RAID Controller’s Auto Import Foreign Configuration setting was not enabled. It is prudent to check if this setting is enabled before creating a new RAID array which implies existing data will be lost. Go to the next step below if the screen above is displayed.

  1. If the following screen is not displayed, push the power button to power OFF and then power ON the server. Select F2 when prompted and to enter the System Setup menu.




  2. From the System Setup menu, for H730P, select Device Settings > Integrated RAID Controller 1: Dell PERC Configuration Utility > Controller Management. For H755/H750 RAID Controller, select Device Settings > RAID Controller in Slot 1: Dell PERC H755/H750 Adaptor Configuration Utility  > Main Menu > Controller Management.








  3. Select Advanced Controller Properties.
  4. Ensure that Auto Import Foreign Configuration is set to Enabled. If enabled, move to next step below. If not set, select Enabled. User must move down to the bottom of screen, and select Apply changes/OK.
  5. Select Back/Finish/Finish/Finish/Yes to exit the menus.
  6. Power OFF the server by pressing and holding the power button located on the right front of the server until the system powers off.
  7. Power ON the server by pressing the power button and let the server boot to see if HDDs will now import.
  8. If the following screen reappears, it indicates that the system failed to import the foreign configuration, which includes the operating system (hypervisor), virtual machines, and other persistent data. This failure typically points to corruption within the drive metadata. A new RAID array must be manually configured, followed by re-imaging of the server. Move to HDD import failure.




  9. If ASP 130 R6.0.x (KVM on RHEL 8.10) has loaded as demonstrated in the screen below, go back to step 13 of this procedure and verify that the Host network configuration was not altered during the import process.




  10. Update the server BIOS/firmware.

    • The process for updating BIOS/FW on an Avaya Dell R660xs is different than on an Avaya Dell R640.

    • Reference the appropriate and latest PSN on support.avaya.com for detailed instructions on updating the BIOS/Firmware.

    As of August 1, 2025:

    PSN027113u - Avaya Solutions Platform 100 Series Dell® R660xs Avaya Certified BIOS/Firmware Update, Version 2

    PSN027112u - Avaya Solutions Platform 100 Series Dell® R640 Avaya Certified BIOS/Firmware Update, Version 16

    • These servers must NOT be updated with BIOS or firmware updates from the vendor’s web site.

    • Only Avaya provided updates can be used. Updating directly from the vendor’s web site will result in an unsupported configuration.

  11. The faulty server returned to Avaya must match the FRU logistics server as far as HDDs, Memory, CPU, PSUs, etc.