RAID 1 is supposed to mirror data to a second disk, why hasn’t this happened?

The explanation

Recently our HP ML115 server’s RAID array became degraded, the array type was a RAID 1 mirror.  The RAID consisted of two 500GB SATA disks and were partitioned as C and E.  When we booted the server we could see in Explorer we now had an F and G drive listed as well as the original C and E drives, the server was now seeing the two individual disks.  F was a replica of C and G was a replica of E.  We noticed that some data was missing from the E drive where the company data is stored, but was on the G drive.  We can boot the server with disk 1 which contains C and E, but we can’t boot using the second disk.  We have since performed tests on the second disk and it is damaged and HP has sent a replacement.  Test performed on the first disk indicate that it is good.

The question

Why should the damaged disk have more complete, up to date data although it is damaged, and the good drive has the missing data?  We believed that data was immediately mirrored across the disks as it was written, supposedly avoiding this discrepancy.  Could someone explain this?

Thanks
grsgAsked:
Who is Participating?
 
DavidPresidentCommented:
Noxcho has it,  your system is now booting to the other disks..
So now you have a really *bad* situation where none of your disks are correct.   About the only thing you can do is manually go through files that have changed between all your disks, and your backup ...  any file that hasn't changed is safe, anything else has to be looked at manually.

This won't be fun.   Sorry.

0
 
DavidPresidentCommented:
Simple, the 2 disks are no longer part of a RAID1, they are now 2 separate drives.  This is not a good situation, as you are no longer running RAID1.   Lose that C/E disk and not only are you left with stale data in F/G, but it probably is corrupted.

Run HP's ACU, assuming it is HP's RAID controller, and repair the broken RAID1 by bringing this disk back into the raidset.
0
 
grsgAuthor Commented:
dlethe, thanks for your reply.  We realise that the disks are no longer part of an array, but why should the data be written to the damaged second disk, the disk containing F and G?  You would think that the data would be written to the first disk containing C and E as these are the original partitions.  I know these questions may seem unecessary, but the MD wants to know the answers.
0
 
noxchoGlobal Support CoordinatorCommented:
Answer could be simple. You are booting from F: and G: drives at the moment.
Get to Windows Disk Management and take screen shot. Post it here.
0
 
grsgAuthor Commented:
Sorry for the late reply guys!  This is very strange; it seems that the server boots using C: but then any data is written to G.  You're right, the data now needs to be manually checked.  Thanks to you both! G
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.