RAID stands for Redundant Array of Independent Disks.

RAID is data storage technology that allows multiple drives to be used together as a single virtual drive for reasons such as fault tolerance, reliability and performance.

There are several different levels of RAID that determine how the data is stored and the level of redundancy achieved across the drives.

Share tech news, updates, or what's on your mind.

Sign up to Post

I have a raid 5 array that has a drive failure. I have an old server that has the same drives in it. Can I put one of the old servers drives into the array? Should I wipe it first?
We have a HP proliant ml350 g5 running 3 drives in raid5 running server 2008 x86, we had a drive in our array go bad . so we swapped the drive with a new one. The server did not rebuild the array like it should have. Instead it thinks a drive on Port 1 Box 1 Bay 2 is a part of the raid which everything else is on Port 2 Box 1 Bay 1 and 3 and WAS 2 . We put the drive in and it now shows Port 2 Box1 Bay2 has a drive in it as unconfigured. i can figure out how to remove the Non existent FAILED drive and add the unconfigured drive back in the RAID so it may rebuild. I have attached the ADUReport.
I have Dell Poweredge R710 with 5x 600GB 15k SAS HDDs with 32GB ram for 150 users.
I am going to configure RAID 10 + 1 HS for dedicated Exchange server.
Should I enable Write Back (default enabled) or keep it disabled as best practice?
on a new machine we discovered performance issues. In the RaidControllers
Logfile we found the following Errors:
TIME = 05-09-2013 17:01:01
LOCALIZED MESSAGE = Controller ID:  0   Link restored on SAS wide port:   0  PHY       =   2

ID = 3795
TIME = 05-09-2013 17:01:01
LOCALIZED MESSAGE = Controller ID:  0   Link lost on SAS wide port:   0  PHY       =   2

ID = 3794
TIME = 05-09-2013 17:00:59
LOCALIZED MESSAGE = Controller ID:  0   Link restored on SAS wide port:   0  PHY       =   2

ID = 3793
TIME = 05-09-2013 17:00:58
LOCALIZED MESSAGE = Controller ID:  0   Link lost on SAS wide port:   0  PHY       =   2

ID = 3792
TIME = 05-09-2013 17:00:50
LOCALIZED MESSAGE = Controller ID:  0   Unexpected sense:   PD       =   Port 0 - 3:1:1Unknown Sense Code,   CDB   =    0x2a 0x00 0x23 0xf4 0xea 0x00 0x00 0x02 0x00 0x00    ,   Sense   =    0x70 0x00 0x0b 0x00 0x00 0x00 0x00 0x18 0x00 0x00 0x00 0x00 0x4b 0x03 0x00 0x00 0x00 0x00 0x00 0x00 0xf5 0x51 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00

ID = 3791
TIME = 05-09-2013 17:00:49
LOCALIZED MESSAGE = Controller ID:  0   Unexpected sense:   PD       =   Port 0 - 3:1:0Unknown Sense Code,   CDB   =    0x2a 0x00 0x23 0xf3 0x32 0x00 0x00 0x02 0x00 0x00    ,   Sense   =    0x70 0x00 0x0b 0x00 0x00 0x00 0x00 0x18 0x00 0x00 0x00 0x00 0x4b 0x03 0x00 0x00…
I am rejigging an HP  server I have - I want to swap out 4 drives from a RAID 10 configuration and put in 4 new drives. The server is running Windows Server 2012 and I have a daily full backup to an attached USB drive.  If this was not RAID - swapping a single drive for a bigger drive - I would just use the Windows 2012 Bare Metal restore procedure. What special considerations, if any, would I need to do this with RAID?  What are the steps?  Any tricks or traps to pay attention to? If somebody out there has done this I would like to hear from you. Thanks in advance.
If I have RAID 5 with 3 disks and if i want to add one more disk to configure 4 disks RAID 5. Can I go ahead and expand?
We are looking at deploying a new server in a site of ours that has no ESX infra.

The app has minimal disk I/O requirements, really its biggest requirement is RAM.

We have the option of:

i. 1 RAID controller set to RAID1 and partition the drives into one C: and one D:

ii. 2 x RAID controllers, one set to RAID1 (OS) and one perhaps set to RAID5/RAID10 for the D:

I know a few years back the standard everyone seemed to go for was RAID1 for OS and RAID5 for data, but with the massive decrease in costs of disk etc, was wondering what the actual benefit of using a separate controller would be?
When creating a software RAID the "normal" way:

Install OS on DISK1 - Change disk type to dynamics - Create mirror to DISK2

then DISK2 obviously is not bootable (mirroring does not mirror the boot record?).

Thus when DISK1 completey fails the client or server cannot be booted any more. How can I prepare DISK2 for this disaster scenario (make it bootable)?

Thanks - Michael

Im looking for a raid driver for my laptop Asus G73s Series.  Please help me.

Trying to expand the RAID 5 array, the new drive is installed and recognized by the system. I am able to set the drive as a spare but am unable to add the drive to the array to increase free space. Have read through HP documentation, have adjusted cache read/write ratio. Everything I am reading says I need a battery backup on the cache module in order to expand the array, can anyone confirm this for me?
I have Dell Poweredge 2900 server and I want to use 4Tb drives. To do so it seems that I need to upgrade the RAID controller. Is this server compatible with PERC H710P, H810, or H710? If so which particular version of the card do I need and how do I do the upgrade? Thank you.
Hi everyone, I want to apologize to all the kind people who replied to my RAID 0 to Single HD.  There was an error.  As it turns out, it is NOT a RAID 0.  It is in fact a RAID 1.  Your advice was great, I just misread the BIOS.  I'm sorry about that.  To recap: I am trying to go from two RAID 1 disks to ONE HARD DRIVE.  Please tell me what to do.  Here is what I have done: First, I have imaged the RAID 1 array.  Then, I went to the BIOS and disabled the RAID.  Now it's set to AHCI.  The other option is "SAT" or something like that.  I heard that AHCI is better?? Then I uninstalled the Intel Matrix app.  When I tried to boot from the new hard drive (with the RAID 1 image on it), it wouldn't boot...just keeps going in loops from bootup.  Since RAID 1 is a mirror, do I need to image ONE disk or both as an array? I imaged the array....Also, It's about 57GB (RAID 1).  I noticed that the image is about 90GB when I tried to run the new HD with the RAID 1 image on it.  That's about twice as the RAID 1 array (maybe because RAID 1 has two disks).   Please tell me what to do in order to make that new SINGLE hard drive bootable.  I want to retain all the data (Vista OS, files, and apps from the RAID 1).  I am new to this and I appreciate your help.
Hi I am acquiring a new server and it will have raid 1 when supplied.

I am then loading the OS.

Question. I have about 70 gb data and do I have to load it on to one volume and the same on the other or will it automatically sync.

I ask this because I was told that synching takes a very long time and affects performance whilst it does.
I need some clarification from a previous post.  Hi everyone, I have a quick question.  Would someone please show me STEP BY STEP how to undo a RAID 0 configuration and putting that data on a SINGLE hard drive (not RAID 0)?  The data is fine, etc etc etc.  All I want to do is to go from 2 disks to ONE disk while retaining data.  Runtime's software doesn't work for me, again.  I can see the two disks in BIOS, as well as in the Intel Matrix, yet I can't get to them. Please recommend another application that can do this task (freeware?? it's not a complicated app).   "My Computer" shows them as ONE disk.  Runtime's app showed them as one disk (Again, and again for some reason.  It would show "HD128:" and not the two disks, not sure why. REPEATEDLY). I need step by step instructions please as I am not familiar with this process. I am not trying to recover any data, just micrate the OS from 2 RAID 0 disks to ONE non-RAID 0 disk while retaining data.  Thanks :)
I have a Dell PowerEdge 2950 with a RAID 5 array.  I use bays 0,1 and 2 for the 3 hard drives that form one virtual disk.  300 Gbytes  in that array (only one virtual disk in the server) are not allocated and I need part of it to extend the current volume in use. File system is NTFS and I am running Windows Server 2003 x64 in that server.
I have no partition management tool to use so my idea is to use diskpart. Problem here is that the partition I want to extend is the system partition and as so, I cannot do it. So I install another hard drive in the server (in the last bay), configured it as a RAID 0 virtual disk, install Windows Server 2003 x64 in it (no updates or anything, just the normal installation which includes SP1) and once I had it up and running I put back the three disks (0,1 and 2 that I took out previously), imported them properly in PERC Configuration Utility and booted the server from the RAID 0.  Now I can see the RAID 5 volume where the real OS for the system resides in Disk Management but it does not show the partition as an NTFS partition. I have not tried to extend it under those circumstances because I do not wan to risk the partition but I do not know why I cannot access that partition as a secondary (or not system partition) in Windows. No need to say that if I browse to My Computer, that partition does not shows up. A last comment is that the partition shows as healthy in Disk Management ( but not file system). The unallocated space in that …
I have OLTP Data base has 3 file groups one of them is a read only file group "historical data about 10 million Record"  contains 1  Physical file

My questions is
what the best RAID LEvel should i place this ReadONLY  filegroup ??
By the way im placing DATA Files ON RAID 5 and LOG File ON RAID 1
So where should i place  the Read Only file group to get a good performance ?
I have a Dell Perc 5i RAID controller that has a mirror and a RAID 10 array on it.  The local LUNs are presented to my ESXi v4.1 as a datastore.  I see that the hardware sensor shows the battery charging and it's been showing this status for a few weeks.  I'm assuming the battery is bad...  Will this state cause the read/write cache to be disabled?  I plan to replace ASAP.
Looks like our SAS 5/iR RAID controller has failed.  If I were to replace the RAID controller with a new one (Same model) will we be able to copy configuration from drives?  Does the SAS 5/iR offer this feature?
We are a small business, 10 users, deploying a new server 2008 or 2012 with Hyper-V.
Hardware is a HP Dl380 G7 with the extra drive cage.
Plan is to set up a several 4 disk Raid 10 arrays to hold the OS. Reason is Raid 10 gives us pretty good security of the data.
I like the idea of Raid 10 having redundancy, but I am questioning whether I am better off setting up several raid 10 arrays or just go with a bigger raid 6 maybe.
I am not sure on the best approach for storage as we move to a virtual environment. The biggest requirement for our server is to provide the storage for our Foxpro data and User Data, I want to create an environment where our techs have lots of storage available for customer documents and site pictures, with server storage for a synced folder backup of everything, and then copy this to a local NAS and then to a offsite NAS
Any good articles on this out there?
Ideas and approaches appreciated.
Hi Experts!

I want to build an insane SSD array.

40+ SSD drives (maybe 64 if I can manage the hardware and physical connectivity space?), SATA III, 500 MB/s (per drive) in RAID-0

Here is what I have so far

Main computer:

Will have 2 PCIE 3.0 x8 slots available.

Surrogate computers (x2 or x4 computer to hold SSD's and power them, connecting to the main computer via external cables [SFF-8088 or SFF-8644)

Will have 1 PCIE 3.0 x8 slot available on each of the 2 or 4 surrogate computers.

I want to produce maximum bandwidth possible. 2x PCIE 3.0 x8 should produce about 9800x2 bandwidth. 19600 MB/s

I need thoughts on hardware.

I am thinking about 4 of these:

Adaptec Series 7 2274700-R 71685

But I am not sure if they would provide the expansion necessary to create the volume and also create the boot drive.

2 raid controllers / expanders (or 1 expander, 1 controller) would be on the main computer, and I would have 2 surrogates or 4 surrogates with 1 raid controller / expander on each surrogate computer.

I would be willing to build 2 more surrogate computers if necessary for a total of 4 surrogates. This would be because I would have to use older cables supporting only 4800 MB/s (SFF 8088 external mini-SAS) rather than the newer faster cables (SFF-8644) capable of at least what the PCI-E 3.0 x8 can handle, 9800 MB/s ...

What I need to know is the hardware capable of handling such a venture.

I prefer LSI or Adaptec, and would prefer to…

I have a Win2k8 x64 server with a RAID5 (3x250GB drives) container for the data partition. Since we are running low on space, I added another physical drive of the same make, model, and firmware. Using the Dell Openmanage admin interface, I added the drive to the container and initiated reconstruction. The reconstruct finished successfully and I show the drive as part of the container with "Used RAID Space: 232GB" and "Available RAID Space: 0.00GB" which is the same as the other drives. The array shows healthy and no errors. However, the available drive space for use in the OS has not changed. I'm curious if I may have missed a step to have the additional space be added to the available usable space for the server.

My HP proliant server running raid0 not able to boot into vmware vmware esxi. When I rebooted, I saw the following message. I did try to hit F2 to accept data loss.... and I was able to boot into vmware esxi. But it happened again today. What is causing the issue? is the first drive physically defective? what should I do next?

Hello Experts,

I use openfiler for my SAN software.

I have 2 more hard drives for my raid 10 array (currently running on 6 drives).

to I pull out all my drives and start to "clean" them using a windows PC and the "clean" command using DISKPART.

After I clean them all and rebuild the array, I find that  when I rebuild the array it still somehow has information from the previous arrays built with those hard drives..

So did a little search on the clean command and find that it just rewrites the "index" file on the hard drives. So it makes sense when I rebuild the array the same I see old info.

anyway so I was looking at the clean all command, but that takes a long time and the one time I tried it, I got some timeout error.

How can I "clean" or "format" these drives so I do not have this problem?

Hopefully there is a method that wont take hours?
Hi Experts

I have a dell Poweredge 840 running server 2003, I have an error showing in open manage, under physical disks, disk 0:0:2 shows up as foreign. The setup is a mirrored raid with a hot spare, however I can only see one other drive, drive 0:0:1 which shows as online. We have 3 drives plugged into slots 1, 2 and 3 should I be seeing drive 0:0:0? how can I tell if its the hot spare showing as foreign or the other disk in the mirror?
They are shown as JBOD it the raid post.  I am wondering if I should clone the drive and put in 2 new drives that are raid, then go from there.

What would you guys recommend to go from windows mirrored to hardware RAID?


RAID stands for Redundant Array of Independent Disks.

RAID is data storage technology that allows multiple drives to be used together as a single virtual drive for reasons such as fault tolerance, reliability and performance.

There are several different levels of RAID that determine how the data is stored and the level of redundancy achieved across the drives.