I have an orange display and an E1810 error showing for Drive 0.
There are 6 146GB SAS drives in the system and a PERC 5/i RAID controller to partition for a 20gb C drive (Windows Server 2003 installed) and 660gb for the data D drive.
I'm assuming that I'm going to have to replace the failing drive but never having had to replace a drive which is part of a RAID array, I'm a little bit cautious so any advice would be appreciated.
Solved! Go to Solution.
There can be a host of reasons why a drive fail that may not nesecistate drive replacement. Different options include checking to see if its in a predictive faulure status meing the drive has exceeded the allotment of errors. Issue can be related to updates as far as firmware on the RAID controller or drives themselves.
As far as replacement, the drive itself is mounted via four screws in the carrier. If the drive has a slow amber light, you can slide the drive out, change the drive and push the carrier with new drive back into the system. With the drives being hot swappable, the drive should kick off a rebuild. If forever reason it doesnt, simply start a manual rebuild via OpenManage or the RAID controller directly, or set as hot spare and that will also kick off a rebuild, thus of course if you didnt have a hotspare already in the unit that took over when this drive fault occured.
I'm waiting for the weekend so that I can power the server down and check what type of drive I need to order - server warranty ran out December 2010 - as I have been told by our supplier that there are two types that could have been originally installed and that one is hotswappable and the other is not.....don't really want to pull a drive and then find its not hotswappable !
Since you have a 2950, there is no confusion as to the status of hot swappable drives. Drive are connected via backplane to the RAID controller. If you physically look at the unit, you will see the drives have a tab in which they can be removed and reinserted.
Just spoken to our supplier who has given me the Dell part numbers that he has been advised fit the Poweredge 2950 :
However the second says that it is for a 2950FS so possibly there is some confusion over the FS ?
The drives currently in the server are indeed caddied so that they are easily removable but it's difficult to say if they are 2.5" or 3.5" because of this.
Regarding the drive size, you can look at the system to identify this.
6 hot-plug 3.5" drives or 8 hot-plug 2.5" drives. If you have 6 slots as you look at the unit, you know they are 3.5" drives
First, you should really have OpenManage Server Administrator installed ... this allows you direct access to the hardware for monitoring and interacting.
Download and run to extract the files, then run C:\Openmanage\windows\setup.exe. Choose Custom and make sure that Storage Management is installed.
Then in Storage, PERC, Connector/Enclosure, Physical Drives, you will be able to see the individual drives - including their sizes and status. Your failed drive should show here either Failed or Online with Predicted Failure/Yes (or missing altogether).
If you want to confirm your drive needs to be replaced, you can run a 2-minute diagnostic on the drive to see (Quick Test):
If it passes, there is a good chance that the failure was simply firmware related - a fluke, you can simply rebuild the drive, then update the system and drive firmware.
If it fails, then replace it. The drives you listed links for one is for a 2.5" drive and one is for a 3.5" drive. Your system will need one or the other. When you get the replacement (if you do need to replace it), make sure you force offline the bad drive (if it is showing Pred Fail), then simply remove the drive and put the new one in. It should begin to rebuild automatically within about 60 seconds (you can monitor this from OpenManage). If it doesn't, you can start the rebuild manually from OpenManage as well by assigning it as a hot spare **. Make sure on a hot-swappable system that you replace the drive "hot" ... don't power it down to replace it.
** There is a chance your drive will show up as Foreign. If it does, you can go to Storage, PERC, Information/Configuration (link at top of the page), Foreign Config/Clear from the dropdown menu.
Both packages are already installed but I've updated the Diag software to the latest version (latest SvrAdmin downloading now)
On Diags, only 5 out of the 6 drives are showing; all of which pass the tests. The front display on the server is now just showing "HDD 0 FAULT" so it appears that the drive has failed. Server at the moment appears to be stable so fingers crossed that we don't have a second failure between now and Dell delivering the replacement 3.5" drive.
You could try reseating the drive - sometimes that is all it takes to re-establish the connection. Sometimes, if the drive has not responded to so many of the controller's requests, the controller will simply ignore it. Reseating also cause the controller to try again. If it shows up after reseating then you can then test it. If you'd rather just leave well-enough alone, then you could test the drive after getting a replacement in and rebuilt. If the drive is still good, then you'd have a spare.
However, even if it does pass, I would strongly consider updating your system and drive firmware, as the causes of many of those fluke failures are addressed in the drive firmware updates.