1

I have a bunch of Dell servers with OpenManage installed.

If I access the RAID manager screens when it goes through the BIOS check, I can view each individual disk and see any issues. Specifically I can see the number of sectors on the disk marked as damaged.

I cannot seem to find this information within the OpenManage screens that you access once the server is up and running (the website on the server).

Does anyone know how I can access this information without bringing the server down and rebooting it? It seems like it should be trivial!

1
  • You can see the information i'm looking for in this screen here flickr.com/photos/marshalljones/3683585968 - The media errors and other errors is the info I want, but without having to reboot and enter the PERC management screens.
    – mjallday
    Jul 3, 2009 at 2:55

4 Answers 4

1

First, make sure you can see the PERC controller and disks under the "Storage" part of the menu on the left side of the screen. If you can't see the controller and drives, then your PERC firmware probably needs to be updated.

Once you have determined that your server storage is visible to OpenManage/ServerAdministrator, you can click on "Storage" (or "System") on the left side of the screen and you should have a "Diagnostics" tab that appears towards the top. From here you can run diagnostic tests on all hardware components in the server including your PERC controller and array disks without rebooting the server.

I have never done this until right now and just finished a test on my 2850. I am using Dell OpenManage Server Administrator v4.4.0 and my PERC 4e/Di is at firmware version 521S.

2
  • Thanks, I had a look and I cannot see the Diagnostics tab you mention. I'm running newer versions of software and firmware (6.1.0 and 5B2D respectively) so I'm guessing that will be the reason. It doesn't sound like something they would remove so with a little more digging I might get somewhere.
    – mjallday
    Aug 5, 2009 at 2:35
  • I just checked my R905 with Dell OM 6.1 and can't see where you would check physical disk consistency either. There is only an option to check the whole virtual disk's consistency, but not the physical disks that make up the virtual. Maybe if it finds a problem with the VD it will give you more specifics then?
    – August
    Aug 5, 2009 at 13:23
0

Start the Server Administrator and in the left pane click on "Storage".

Note that this will only show anything if you ticked the Storage Management option when you installed the OpenManage stuff. Your tag implies you have a 2850, presumably with a Perc4e, so you should be able to see your controller config there. With SAS5/iR controllers older versions of Openmanage don't support this controller and don't show anything.

Assuming your Openmanage shows the RAID controller navigate to the virtual disks and click on the virtual disk name to see all the drives that are in the array.

If you don't see anything under the Storage heading just download and install the latest version of OpenManage and make sure Storage Management is included as an installation option. NB uninstalling the old version of OpenManage will probably need a reboot.

JR

2
  • I've looked into these screens before, they have a bunch of info (Speed, capacity, failure predicted etc) but no mention of the number of bad blocks. Screen shot - flickr.com/photos/marshalljones/3652807751 - is this the screen you mean? Thanks for the help!
    – mjallday
    Jun 23, 2009 at 8:07
  • Bad blocks? As far as I know the RAID BIOS doesn't show bad blocks either. SCSI disks shouldn't show any bad blocks anyhow, as they dynamically remap any sectors deemed to be at risk. If you look in the OpenManage logs they will contain any errors or warnings from the RAID controller. I forget which log they usually appear in. Jun 23, 2009 at 8:22
0

Instead of OpenManage you might want to try Dell's OpenManage Array Manager. This provides better access to the RAID controllers.

It may have become unsupported in the last few years, but it might do the trick.

Also, this information should just be SMART ... in theory, any SMART monitoring tool should show you this info (regardless of the array). Try smartmontools if you're running linux. And wikipedia has a listing of SMART monitoring applications.

1
  • doesn't work with the last two generations of servers. all the finctionalityhas been moved over to OMSA
    – dyasny
    Sep 28, 2009 at 10:03
0

You will not see bad block numbers anywhere. What you will see is media error counts, which amounts up to a predictive failure error.

You must log in to answer this question.

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