3

My client is getting a used HP Proliant DL360 G6 server and I want to make sure that it's safe to use this server further in production.

I've tested RAM using GoldMemory (results are OK), and now I'm lost searching for a way to test hard drives. At this point I know that neither Viktoria nor MHDD can see the drives, partially because P410i doesn't run in so called JBOD ("Just a Bunch Of Disks", I guess) mode.

So, how do I test these hard drives? Thanks in advance for any help.

Improve this question
5
  • 1
    This process of testing the disks is unnecessary.
    – ewwhite
    Jun 20, 2016 at 16:59
  • @ewwhite but how can I make sure that these used disks are OK to use them further in production environment?
    – real_sm
    Jun 24, 2016 at 9:14
  • Read the RAID controller status at boot (it will tell you the drive state), and continue to monitor once you've installed the OS. I don't see the problem. You would use RAID to protect against individual disk failures the same way you would with new disks.
    – ewwhite
    Jun 24, 2016 at 9:15
  • @ewwhite OK, thanks. I suppose, it's the "enterprise-way" of managing servers, and I will try to get used to it, though we are short of money (that's why company bought used server).
    – real_sm
    Jun 24, 2016 at 9:25
  • This is HP equipment. It's higher quality. See my post on automated testing.
    – ewwhite
    Jun 24, 2016 at 9:26

4 Answers 4

Reset to default
1

There's no reason to "test" the drive. The controller will tell you whether the drive is healthy or not at POST during the RAID controller initialization.

Treat this like any other server and plan for component failure. Make sure the drives are in an acceptable RAID level and that you monitor the system's health from your operating system.

Also see: Automated hardware testing of HP servers?

Improve this answer
1
  • OK, I'm just not used to HDD self-test in servers. Usually, with desktops, when an error message shows at POST, it means that the drive is not usable at all.
    – real_sm
    Jun 24, 2016 at 9:23
3

Using the HP SmartArray utilities, you can see the SMART information for the drives.

Improve this answer
2
  • 1
    I'm sorry, but could you please kindly point me to where can I see that info in HP SmartArray utilities? There is a lot of stuff in diagnostics log, but I can't see SMART info (for instance, how many hours did the drives work for). Thank you.
    – real_sm
    Jun 24, 2016 at 9:17
  • youtu.be/3-4ogWgIslY - video reference
    – andreikashin
    Oct 11, 2020 at 14:54
1

You can use HP Service Pack for Proliant DVD. There is a bunch of tests on it, to test wide variety of HP server subsystems. Just boot from it, and you'll see all the options.

It is available on HP site via support contract http://h17007.www1.hp.com/us/en/enterprise/servers/products/service_pack/spp/index.aspx

Or you can search through internet, no doubt you will find that you need.

P.S. Also keep in mind, that newer versions of SPP may not support older servers hardware, therefore you may need to find older one.

Improve this answer
1
  • I tried that DVD, the tests are the same as in Windows HP Smartarray utilites, but I can't find any surface test or SMART info there.
    – real_sm
    Jun 24, 2016 at 9:18
0

If you want to do burn-in on the drive, there are a few things you can do. For system utilities, smartctl and hdparm have low level test modes. bonnie++ is a higher level stress test.
Here is a thread with a few examples of stress testing drives. https://unix.stackexchange.com/questions/72563/is-there-a-good-drive-torture-test-tool

Improve this answer
1
  • I just want to test the surface, like in MHDD.
    – real_sm
    Jun 24, 2016 at 9:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .