topic BL860c Boot Disks Synchronization in BladeSystem - General

BL860c Boot Disks Synchronization

rianui — Wed, 09 May 2012 16:11:05 GMT

Hi Gurus,

I have a bl860c running Redhat Linux. One RAID 1+0 boot drive failed and so I replaced it. I got into the EFI shell and called up the RAID settings. I saw that the disks were synching and left restarted the OS. However when I check the new drive after more than eight hours of synching it still blinks as though it was just inserted. When I tried to boot with it alone, there was no OS on it.

How long does synchronization take? Also, how do I ascertain that the two disks are synching?

Regards,

Rianui

Re: BL860c Boot Disks Synchronization

Torsten. — Wed, 09 May 2012 17:56:38 GMT

It depends on the disk size and work load ...

I have seen several times a 300GB disk inside a BL860c i2 syncing about 24 hours.

Re: BL860c Boot Disks Synchronization

rianui — Thu, 10 May 2012 10:45:02 GMT

Thanks Torsten,

It's still blinking as though it was just inserted....over 12 hours later.

I also keep getting the following errors:

1. Kernel: panic on journal commit I/O error
2. scsi: 0:1:1:0 rejecting I/O to dead device

The server is connected to an MSA2312fc storage box and I suspect that the I/O errors are coming from there. Controller B on the MSA is showing faulted (Supercapacitor failure. Status: cell 4 charging voltage too is high (26), state: failure (12) (p2: 2546, p3:0)

According to an HP advisory this error points to a firmware release and they advise a downgrade. Mine is on the firmware they say it should be at but I still get this error.

I must absolutely identify where the I/O error is coming from (it's hanging the production server) and not just suspect the storage controller because even after starting the storage with only the healthy controller, the error came up again :-(

NB: The internal disk that I added to mirror the OS showed this path

SCSI: 0:0:1:0 generic sg25 type0

Regards,

Rianui

Re: BL860c Boot Disks Synchronization

rianui — Wed, 16 May 2012 13:36:19 GMT

I figured out that both root disks were messed up. After logging a call with HP they said the same and added that the SCSI disk controller should be replaced as well. I will replace these components and configure a new RAID volume then install Linux.

Regards,

Rianui