-2

I have 3 of the aforementioned servers, and when I netinstall centos on them, the installer recognizes the raid configuration and the logical disks and installs fine. Upon reboot after installation, the server reports that it is attempting to boot from drive C: and never gets past that.

I've used CentOS on dozens of servers before and never really had an issue, but most of those were custom ordered to be compatible with CentOS. These were dropped in my lap and I was told to 'make it work'.

The servers are Generation 8, and the have the HP Dynamic Smart Array B320i RAID controller. I haven't had any time to do research on the issue myself, I just wanted to drop SF a line to see if anyone else had noticed a similar issue. The system has six physical SSDs split into three logical drives, slots 1 and 2 are a RAID 1 for the OS. Slots 3-5 are a RAID 5 for data, and slot 6 is a single drive for logs.

I tried switching the controller to AHCI which I read on some HP Knowledgebase, but that didn't seem to have an effect.

Any help is greatly appreciated.

spuy767
  • 203
  • 3
  • 4
  • 11
  • ** ... Check the boot order in the BIOS. While you're at it, you should probably make sure your Linux installation is actually the whole of your RAID array, and you don't have part of a Windows install hanging around on the disk. – HopelessN00b Jan 21 '15 at 01:38

1 Answers1

0

HP servers get updated from time to time and it is important to state the generation of the server. With that information, you could lookup whether the server is certified for the upstream product from Red Hat.

Also with the server type you could find out, what raid controller is used and what driver might be needed.

As far as I remember, there are different Proliant 360 raid controllers used. One creates well known /dev/cciss/c0d0 devices and is well supported under Linux. Another raid controller that is usually used for cheaper builds is software based and that might be the cause for your troubles.

To fix this issue, I would boot in rescue mode from the CentOS installation dvd. Then identify your installation (as the device was found during installation, the driver seems to be available for the anaconda installer).

When you chroot to your installed root filesystem, I would download the support pack for proliant and would install all drivers and rebuild the initial ramdisk.

I would also check the boot components (grub, kernel, initrd).

Reiner Rottmann
  • 633
  • 1
  • 7
  • 19