Link to home
Start Free TrialLog in
Avatar of benc007
benc007Flag for United States of America

asked on

RAID 5 with Three Hard Drives

I have RAID 5 with Three Hard Drives.  One of the drives is failing on an important server.   In RAID I see:

Array Disk 0:0   Failed    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity
Array Disk 0:1   Online    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity
Array Disk 0:2   Online    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity

I think the three hard drives are 147 GB each but I'm not sure.  They are made by Seagate.

I have an extra Fujitsu SCSI hard drive that is 147GB ... it is a different brand.  I want to pull this from another live server, and install it as "Array Disk 0:0" on this server.  What will happen?
Avatar of pjam
pjam
Flag of United States of America image

Your array should configure it as 136.63 and off you go.
raid 5 will always make a larger drive the same size as the smaller, so if it was say 300GB it would use it as 136.63 also.
Avatar of Member_2_231077
Member_2_231077

They are already the same size, 136.6 GiB is 147 GB. One program is reporting the size in powers of base10 and the other in powers of base2. 1MB = 1000B, 1MiB = 1024B. Disk manufacturers (and OSI) use powers of base 10. https://en.wikipedia.org/wiki/Gibibyte

If it is a Dell disk then be sure to use a Dell  branded replacement, same for IBM. If it and the others are generic then use a generic one. As per above you can be safe by shoving a higher capacity one in as long as it is the same technology.
Avatar of Seth Simmons
What will happen?

it should start rebuilding automatically
what server and raid controller are we working with here?  depending on what is being used, could require manual intervention
All the time one disk is failed in a RAID-5 setup your data is at risk of a second disk failure so you need to get the third disk back in action ASAP.

We do not recommend the use of RAID-5 anymore with the caveat that this really only applies to disk over 750GB, so does not apply in your case.
Sometimes this can happen if our battery cache is bad.  Do you have something like array diagnostics?
You did not mention server model or array hardware, that will help.
Avatar of benc007

ASKER

This is a PowerEdge 1750 with PERC 4/DI.

pjam - I have Dell OpenManager.  How do I run array diagnostics?
Avatar of benc007

ASKER

How can I see if the battery cache is bad?
It's now a week later, I hope you have a backup of your data, if not do one before you do anything else!
The manual is here:
http://downloads.dell.com/manuals/all-products/esuprt_ser_stor_net/esuprt_poweredge/poweredge-1750_owner%27s%20manual_en-us.pdf
look at page 43, "Running the System Diagnostics".
Does your server have Integrated or physical Raid card?
Avatar of benc007

ASKER

RE: This is a PowerEdge 1750 with PERC 4/DI.

pjam
1) How can I see if the battery cache is bad?  Are you referring to a dead RAID battery?
2) How can I see if my server has integrated or physical RAID card?
It doesn't matter if the battery is good or bad, you just have to replace the failed disk. If the old one comes out of another Dell it will have metadata on it so in OMSA you would have to erase foreign configuration under controller tasks then set it as a hot spare after which it should start to rebuild the logical disk.
Avatar of benc007

ASKER

andyalder - I have another "working" hard drive from a different PowerEdge 1750 server, which has some old data on it.

If I hot swap this working hard drive with the current failed hard drive, will Windows automatically rebuild this hard drive?

RE: in OMSA you would have to erase foreign configuration under controller tasks then set it as a hot spare after which it should start to rebuild the logical disk.
Do I have to do this manually before hot swapping the failed drive with a working hard drive?  How do I do this exactly?
DELL PERCs do not rebuild automatically onto a foreign disk,

If the old 1750 does not have any data you want to keep then you can boot that, enter the BIOS utility <ctrl>R and erase the disk configuration, that will clear the metadata and then when you hot-swap the disk into this server.

If you can't use the donor server to erase the metadata then on this one you can go into OMSA, storage, PERC, click information/configuration link and there is an option to clear foreign configuration there. Once the foreign configuration is cleared it may start to rebuild automatically or you may have to assign it as a hot spare under PERC-logical disks.
Avatar of benc007

ASKER

RE: DELL PERCs do not rebuild automatically onto a foreign disk
-> Are you sure about this?  This Dell 1750 had a failed drive before, and when I replaced it with same identical drive (brand, size and serial number) that had NO data on it, I was able to hot swap the failed drive and the server rebuilt the failed drive and restored RAID 5.  This was an empty drive though.

The other 1750 good hard drive is not the same brand and serial number, but it has data on it.  This 1750 is also set up in RAID 5, which as data that I need.  I don't think I can erase the disk configuration on just ONE of the three drives.
"Foreign" means a disk with metadata on it, not the same as a clean new disk. If you need the data in the other server then you would have to do it with OSMA on this server as I outlined earlier.

Don't like the idea of taking a disk out of a working server that you need the data on.
Avatar of benc007

ASKER

RE: Don't like the idea of taking a disk out of a working server that you need the data on.
I don't either, but it's not a critical server.  If I remove 1 of the 3 drives, it will lose RAID 5, but the server will still work right?

For the donor 1750, if I remove 2 of the 3 drives, will the server boot with just one drive?  

After rebooting in Windows Server 2000, how can I remove all metadata on it?
If you remove one of the drives the server should keep working, if you remove two it dies.

I've already told you how to remove the metadata on the foreign drive using OpenManage Server Administrator.
Avatar of benc007

ASKER

I will try to buy a used Seagate 146GB SCSI 15K drive.  Does this drive need to be the same model number / serial number as the other 2 drives?

I don't have another 1750 to see if this drive has meta data or not.  If it does have meta data, and I hot swap into my 1750, what could happen?
Don't buy a generic Seagate disk, get the Dell one , there should be a barcode label on it with a 5 character part number such as GC828 in the string.

I've already told you what will happen if you plug it in - nothing until you erase the foreign config with OMSA.
Avatar of benc007

ASKER

RE: I've already told you what will happen if you plug it in - nothing until you erase the foreign config with OMSA.
I am buying a 146 GB SCSI 15K drive.  The seller says there's the drive has been WIPED CLEAN, which I think it means there is NO METADATA on it, right?

I have RAID 5 with Three Hard Drives.  One of the drives is failing on an important server.   In RAID I see:
Array Disk 0:0   Failed    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity
Array Disk 0:1   Online    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity
Array Disk 0:2   Online    SCSI    ProductID   Rev0003   Seagate   136.63 GB capacity

RE: Don't buy a generic Seagate disk, get the Dell one , there should be a barcode label on it with a 5 character part number such as GC828 in the string.
My hard drives are made by Seagate, why would I get a Dell branded drive instead?
Avatar of benc007

ASKER

Can I just rebuild the one failed drive?  And then put it online?  

Please see http://prntscr.com/evbuv6
"My hard drives are made by Seagate, why would I get a Dell branded drive instead?"

Because, the Dell branded drives, will have code page tweaks and possibly Dell specific firmware on board, and the controller may do different things with generic drives, than it will with branded drives.
ASKER CERTIFIED SOLUTION
Avatar of Member_2_231077
Member_2_231077

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of benc007

ASKER

I bought a used drive that was taken from production and is supposed to be "fully tested and reformatted to NTFS for Windows OS".

1) If I hot swap this drive with the failed drive on my Window 2000 Server, while the server is running, will this drive rebuild without affecting the websites hosted on this server in real-time?  There are live users on these websites.

2) I have another PowerEdge 1750 that has 3 hard drives also set up in RAID 5.  If I remove these 3 drives, and put the SINGLE used replacement drive and power on the server, how can I test this used drive?  

After testing this used drive, when I put the 3 hard drives back in, will everything run the same as before ... or will I have to make some re-configurations?
136 GB SCSI?  Time to replace the server.  All 3 disks should be pretty old.  If you're replacing one now, the 2nd one may be ready to fail soon.  Back up your data.

Unless this is some critical server with special software, it's time to replace the unit.  136 GB SCSI is ancient.  The 1st disk failed already.  The other 2 are likely the same age.  They'll be failing soon too.  Back up the data.  Upgrade the old worthless unit.  It's using more electricity monthly than an Intel NUC or a laptop and either of those are currently more powerful than your old Dell PowerEdge 1750.
Avatar of benc007

ASKER

Thank you so much for your help Andy!

Everything worked out  =)