Help Center
Home Server & Cloud Infrastructure Dedicated Server

Setting up RAID 1, 5, or 6 in the Rescue System on servers with hardware controllers and netboot.xyz images

Valid for Dedicated Servers and Bare Metal Servers with the netboot.xyz image.

In this guide, you will learn how to boot your Dedicated Server or Bare Metal Server into the Rescue System and configure a hardware RAID (RAID 1, 5, or 6) using the command-line tools storcli or perccli. This is necessary to manually adjust the hardware configuration before installing the operating system.

If you order a custom-configurable Dedicated Server or Bare Metal Server with hardware RAID 1 and hardware RAID 5, and install a netboot.xyz image on it, you must use the command-line tools storcli or perccli to configure a RAID 1 on the primary data storage and a RAID 5 or RAID 6 on the additional HDD or SSD data storage.

Notes

  • Creating a hardware RAID requires very good knowledge of server administration with Linux.
  • The creation and configuration of a hardware RAID is not included in the support scope. An overview of the included services can be found on the following page: Scope of support for server products

Please Note

  • Creating a new RAID will delete all existing data on the target hard drives.
  • Ensure you create a backup before making any changes to an existing RAID array.

Boot the server into the Rescue System

Before you can make changes to the RAID controller, the server must be started in a neutral environment.

  • Log in to your IONOS account.

  • Click on Menu > Servers & Cloud in the title bar. 
    Optional: If you have multiple server contracts, select the desired contract.

  • In the Infrastructure > Servers section, click on the desired server.
  • Click Actions > Restart. The Restart Server window opens.
  • Activate the Rescue System method.
  • If you have installed the netboot.xyz image, select the Netboot.xyz latest rescue system. If you have installed a different image, select the Rescue system (Linux) rescue system.
  • Click Yes. The server will boot into the Linux rescue system.
  • To log in to the Linux rescue system on your server, establish an encrypted network connection to your server via SSH using the credentials displayed.

Identifying the hardware controller and drives

After connecting to the Rescue System via SSH, you need to determine which controller and tools are available.

Check which tool is available on your system. Enter the following commands one after the other to query the version:

  • storcli64 -v (for common Broadcom/LSI controllers)
  • storcli2 -v (for the latest generation of Tri-Mode controllers (NVMe/SAS/SATA))
  • perccli64 -v (for older generations of Dell PowerEdge RAID controllers)
  • perccli2 -v (for the newest Dell generation, e.g., GraniteRapids/Blackwell systems)
Display the controller ID and status of existing physical drives

Note

In this guide, we use storcli64 as an example. Replace this command with storcli2, perccli64, or perccli2, depending on your system.

  • Display the installed controllers:

    storcli64 /call show

  • Note the controller ID (usually 0).

  • Check the status of the existing physical drives (Physical Drives - PD):

    storcli64 /c0/eALL/sALL show all

  • Pay attention to the drive status. To create a RAID, the drives must be in the Unconfigured Good (UGood) status.
  • Note the Enclosure ID (e.g., 16 or 252), the slot number of the hard drive, and the serial number.

Note

When replacing a drive, always compare the serial number with the information provided by IONOS Customer Support.

Create a hardware RAID

Select the appropriate RAID level based on your requirements. The general syntax is:

storcli64 /cx add vd type=raidX drives=EnclosureID:SlotID,SlotID…

Creating RAID 1 for the operating system

A RAID 1 requires at least 2 hard drives. It provides full redundancy. The storage capacity of the array is at most equal to the capacity of the smallest data drive.

To create a RAID 1 array, run the following command:

storcli64 /c0 add vd r1 drives=ENCLOSURE:SLOT1,ENCLOSURE:SLOT2

In this example, a RAID 1 is created from two hard drives located in slot 4 and 6 of enclosure 16:

storcli64 /c0 add vd r1 drives=16:4,16:6

Creating RAID 5

RAID 5 requires at least 3 hard drives and uses striping with distributed parity. It provides increased fault tolerance for the failure of a single drive through parity. Overall, RAID 5 offers a good price-to-performance ratio regarding redundancy and storage optimization.

To create a RAID 5 array, run the following command:

storcli64 /c0 add vd r5 drives=ENCLOSURE:SLOT_NUMBER1-SLOT_NUMBER3

Example:

storcli64 /c0 add vd r5 drives=8:0-2

Alternatively, you can use the following command:

storcli64 /c0 add vd r5 drives=ENCLOSURE:SLOT_NUMBER1,SLOT_NUMBER2,SLOT_NUMBER3

Example:

storcli64 /c0 add vd r5 drives=8:0,1,2

Creating RAID 6

RAID 6 requires at least 4 hard drives. It uses block-level striping with double parity. The major advantage over RAID 5 is increased fault tolerance: the system remains operational even if two hard drives fail simultaneously.

With this command, you create a RAID 6 with 8 hard drives located across two enclosures:

storcli64 /c0 add vd r6 drives=ENCLOSURE1:SLOT_START1-SLOT_END1,ENCLOSURE2:SLOT_START2-SLOT_END2

Example:

storcli64 /c0 add vd r6 drives=17:0-3,33:0-3

With this command, you create a RAID 6 with 12 hard drives located across two enclosures:

storcli64 /c0 add vd r6 drives=<ENCLOSURE1>:0-5,<ENCLOSURE2>:0-5

Example:

storcli64 /c0 add vd r6 drives=17:0-5,33:0-5

With this command, you create a RAID 6 with 12 hard drives located across more than two enclosures:

storcli64 /c0 add vd r6 drives=<ENCLOSURE1>:<SLOT_START1>-<SLOT_END1>,<ENCLOSURE2>:<SLOT_START2>-<SLOT_END2>,<ENCLOSURE3>:<SLOT_START3>-<SLOT_END3>

Example:

storcli64 /c0 add vd r6 drives=6:0-3,8:0-3,16:0-3

Verify the configuration

After creation, the new virtual drive (Virtual Drive - VD) must be verified. To display information about the new virtual drive, enter the following command:

storcli64 /c0/v0 show

Set boot capability

To ensure that the BIOS/UEFI recognizes the primary RAID 1 as the boot drive, the corresponding virtual drive (Virtual Drive) must be marked as a boot drive on the RAID controller.

If both a RAID 1 for the primary data storage and a RAID 6 for the secondary data storage are used on the server, you must ensure that the RAID 1 is set as the boot drive.

To do this, adapt and run the following command:

storcli64 /c<CONTROLLER_ID>/v<VIRTUAL_DRIVE_ID> set bootdrive=on

Example:

storcli64 /c0/v0 set bootdrive=on

In this example, the virtual drive /c0/v0 is set as the boot drive.

Ensure that v0 is actually the primary RAID 1 for the operating system and not the secondary RAID 6.

Special considerations for UEFI and BIOS

Modern Bare Metal Servers and Dedicated Servers use UEFI by default.

GPT partitioning: If your created RAID is larger than 2 TB, you must use a GPT partition table during the subsequent operating system installation, as BIOS/MBR does not support this size.

Install the operating system via netboot.xyz

Once the hardware layer is configured, the installation of the operating system follows:

  • Once the initialization is complete, you can exit the Rescue System.
  • Go back to the Cloud Panel.
  • In the Cloud Panel, click Infrastructure > Servers and select the desired server.
  • Click Actions > Reinstall Image.
  • Install the netboot.xyz image.
  • As soon as the installation is complete, click the Console button. To establish a connection with the VNC console, select the VNC Console. The VNC console will then open in a new tab, and the connection to the server will be established. This may take a moment.
  • Select the option Continue with Self installation Here. The menu will load.
  • In the Distributions section, select the installation method GNU/Linux.
  • Select the desired Linux distribution and follow the instructions.

Content