OM reporting failed drives

Question

We have just setup two new servers with san in the following configuration: Dell R620 x2 with 2012 hyper configured in a cluster connected to MD3220 via SAS with one SAS cable on each of the 2 onboard controllers.

Dell Openmanage is reporting failed two drives on 6Gbps SAS HBA controller 0 and 1. Upon checking the MD3220 no drives have failed.

Below is an extract from the logs

ID 0:7 Status Critical Name Physical Device 0:7 State Failed Execute Bus Protocol SAS Media HDD Revision 0784 Capacity 0.00GB Used RAID Disk Space 0.00GB Vendor ID DELL Product ID MD32xx Serial No. Negotiated Speed 6.00 Gbps SAS Address 590B11C406C46904

Has anybody encountered errors similar to these?

Cheers,

Dave

dave_mc15 · Accepted Answer

I have discussed our issue with Dell support last month, Dell support escalated the issue through there internal support to product support SME and verified that there is a bug in openmanage version 7.3. Open manage shouldnt be able to show any drives attached to the SAS HBA but for some reason it was trying too.

This issue should be fixed in the next release such is due March 2014.

As long as the MDSM is reporting the disks as good then all should be ok.

DELL-Sam L · Answer

Hello Dave,

What I would recommended is that you install MDSM (Modular Disk Storage Manger) as that is what you will need to manage the MD3220. Here is a link to download the software. www.dell.com/.../powervault-md3220 This is an ISO so you will need to burn it to a dvd to install. Once installed then you can check it there & see if it is reporting the same error. If you are getting any error you can run a support bundle from the support tab in MDSM so that we can review it to see what is causing the error. I will send you a private message that you can send the logs to.

Please let us know if you have any other questions.

RC-1 · Answer

Hello Sam,

it seems like I'm havimg a similar problem, I have 2 powerEdge R710 connected to a MD3200 and suddenly both R710 are reporting critical error with drives 6Gbps SAS HBA, Physical device on connector 0 1:3; and conector 1 1:3 and 0:7, I had a chat with Dell agent and after providing him the generated DSET report on one of the R710, he says there's no problem on it, that he sees that the problem must be on MD3220, but on MDSM the array looks perfect, no error or a single alert, can you please help me?

Thanks in advance

Dev Mgr · Answer

Did you install MDSM (or at least the host software) on the server(s) that use the HBA to connect to the MD3200? This is a requirement (if you're running Windows (including Hyper-V server) or Linux on the physical server.

Dev Mgr · Answer

Dave_md15 posted that he got info from Dell support that it's a bug with OpenManage 7.3. Is this the version of OpenManage that you are running?

Podini Holding · Answer

I have the same Problem on 4 x PowerEdge R710 with Windows 2012 R2 Hyper-V Server connected to a MD3200. Any News on this?

Podini Holding · Answer

Yes Version 7.3.0.

Dev Mgr · Answer

Then you are running into the same bug most likely. Either upgrade to a newer OMSA (if available), or downgrade to an older version till a newer version is available.

Marian Vulpe · Answer

Thanks for the topic. We've had the exact same issue, solved by upgrading OMSA from 7.3 to 7.4.

JM99 · Answer

Okay, I have slightly off-topic, but somewhat (maybe?) related - I am on 2008 R2, just put on MDSM version 4.1.0.88 (it is recommended, even though it is technically for 3200/3600 & higher; Dell states that it will manage all MDxxxx - and it does seem to do that); I get the same error; have yet to upgrade to 7.4 OMSA; but, would any of this also prevent the storage from being presented to the OS? I think I have followed the steps for setting up MD3000 (not the "md3000i," just md3000) - completely "in-band" I don't care about any Ethernet or iSCSI, etc.; and yet, the os does not yet see the drives - but sees 2 failed drives, as you mention above. Also, in putting on the latest MDSM, it has a SAS 5/E driver but, if I try to load that, it tells me "Not the correct drive for your platform." Server hardware is R900. I chose RAID10, one hot spare, and "Windows MSCS clustered - single-path" (tried to 2008 clustered and tried 2008 standalone also) - I'll double-check cabling - maybe I hosed the physical cabling config.

Also, does the "split-mode," "single-mode" switch on the MD3000 make any difference at all?

dave_mc15 · Answer

Upgrading to 7.4 fixed the original issue we had in our customers environment.

In regards to the drives not been seen by OS, did you create mapping from the MD storage to allow the hosts to access the LUNs created.

PowerVault

OM reporting failed drives

Was this post helpful?