Do you wonder if your IT business is truly profitable or if you should raise your prices? Learn how to calculate your overhead burden using our free interactive tool and use it to determine the right price for your IT services. Start calculating Now!
HP Proliant DL380 G4 RAID LED explanation
The server had some event, when I got to site 4 of the 6 drives where flashing amber. The monitor just had a flashing "-" at the top left.
I rebooted and it claimed it needed to rebuild one of the drives, so I let it. Now all the drives are green, except 1 or the 3 RAID5 array drives that is flashing green. Monitor is showing the same "-".
While I was composing this, the flashing drive is now normal, but the single driver and the 2 in a RAID1 are now showing solid amber [X].
For clarity:
Drives 0/1/2 are RAID5 (3x146GB SCSI 15k).
Drive 3 is standalone (36GB).
Drives 4/5 are RAID1 (2x300GB SCSI 10k)
The RAID5 is the Windows Server booting array.
What can I deduce from the above?
I rebooted and it claimed it needed to rebuild one of the drives, so I let it. Now all the drives are green, except 1 or the 3 RAID5 array drives that is flashing green. Monitor is showing the same "-".
While I was composing this, the flashing drive is now normal, but the single driver and the 2 in a RAID1 are now showing solid amber [X].
For clarity:
Drives 0/1/2 are RAID5 (3x146GB SCSI 15k).
Drive 3 is standalone (36GB).
Drives 4/5 are RAID1 (2x300GB SCSI 10k)
The RAID5 is the Windows Server booting array.
What can I deduce from the above?
Do more with
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
The flashing green sounds like it was rebuilding.
The amber X suggests failed disks but need to know other LED status. http://h20565.www2.hpe.com/hpsc/doc/public/display?sp4ts.oid=397634&ac.admitted=1404344901177.876444892.492883150&docId=emr_na-c00357539-6&lang=en-us&cc=us&docLocale=en_US#N10671 has the disk error LEDs table.
Much easier to look in the ACU and see what that says than to look at the LEDs, also you can use the ADU to generate a report you can upload here as an attachment. Under Windows ACU and ADU are under start->hp system tools.
The amber X suggests failed disks but need to know other LED status. http://h20565.www2.hpe.com/hpsc/doc/public/display?sp4ts.oid=397634&ac.admitted=1404344901177.876444892.492883150&docId=emr_na-c00357539-6&lang=en-us&cc=us&docLocale=en_US#N10671 has the disk error LEDs table.
Much easier to look in the ACU and see what that says than to look at the LEDs, also you can use the ADU to generate a report you can upload here as an attachment. Under Windows ACU and ADU are under start->hp system tools.
Thanks for your feedback. Since I posted I tried to boot and the drives looked ok, however the 3rd drive in the RAID5 wanted to rebuild again, so I let it. It finished and I rebooted, but now all drives except the drive 0 of the RAID5, so obviously it doesn't boot because 2 out of 3 of the RAID5 are Amber.
Given the up and down of the different drives, could it be a controller issue rather than a drive issue?
Given the up and down of the different drives, could it be a controller issue rather than a drive issue?
Could be the controller but equally well the drive in the RAID 5 that keeps rebuilding could be causing bus faults so I would remove that and try booting again.
Ok, just been to site and tried removing the dodgy drive, but it still won't boot. The 3 drives in the RAID5 array only have LED 2 lit each , the 2 in RAID1 have 1 & 2 LEDs lit and the standalone drive is Amber. On that basis I tried removing the Amber drive and it still doesn't boot.
What message is there on the screen when the array controller initializes? It was the drive that kept rebuilding from the RAID 5 group I meant to remove.
It certainly could be a fault with the controller, it's integrated on the mobo but you can use a Smart Array 642 in a PCI slot instead, it will read the metadata.
It certainly could be a fault with the controller, it's integrated on the mobo but you can use a Smart Array 642 in a PCI slot instead, it will read the metadata.
It's a Smart Array 6i controller. I'm not sure, but think that's a separate card. If it is faulty, is there a way to replace without losing the RAID?
6i is integrated onto the motherboard but you can put a 641 642 or 6402 in a PCI slot if there's one free. The replacement card will read the configuration from the disks so no worries about swapping mobo or replacing it with a Smart Array PCI card. The cache DIMM on the 6i is on the motherboard too, under the PCI riser cage.
Do you have one or two SCSI cables coming from the system board to the disk backplane?
Do you have one or two SCSI cables coming from the system board to the disk backplane?
Ok, together with HP Support we managed to get it running again. Simplistically there was an F1 an F2 option. I shied away from the F2 option because it hinted at data loss, but all it did was re-enable the disabled drives. We are back in Windows and breathing a sigh of relief.
Thanks for your contribution!
Thanks for your contribution!
Yay!
I did ask what was the message, the F1/F2 prompt is a common one if you re-introduce a disk back into an array like your dodgy one in the RAID5. F1/F2 data loss message is because it is going to overwrite what's on the disk with parity/data during rebuild, it's pretty misleading in that that's what you want it to do. Say you took a disk out and put it in another machine and then wrote some important data on it, then stuck it back in this one then the data loss warning would be valid, but 99.9% of the time it's just annoying. They've changed the firmware for later controllers so it defaults to re-enabling the logical drives.
May still be a good idea to get an ADU report and post it here, the disk statistics at the bottom of the report can reveal problems that the LEDs and ACU don't show.
I did ask what was the message, the F1/F2 prompt is a common one if you re-introduce a disk back into an array like your dodgy one in the RAID5. F1/F2 data loss message is because it is going to overwrite what's on the disk with parity/data during rebuild, it's pretty misleading in that that's what you want it to do. Say you took a disk out and put it in another machine and then wrote some important data on it, then stuck it back in this one then the data loss warning would be valid, but 99.9% of the time it's just annoying. They've changed the firmware for later controllers so it defaults to re-enabling the logical drives.
May still be a good idea to get an ADU report and post it here, the disk statistics at the bottom of the report can reveal problems that the LEDs and ACU don't show.
Premium Content
You need an Expert Office subscription to comment.Start Free Trial