PE2850: PERC 4e/DI Controller 0, Array Disk 0:0 failed

Yes, you have done something wrong - not to be harsh - but let me be honest and explain:

• Never remove a hot-swap drive while the system is off; if it is hot-swap, swap it hot.
• Never do anything with a RAID controller without understanding 1) the basics of RAID, and 2) the particulars of the controller you are working with.

Support in the States/Canada is always free, 24x7, regardless of warranty status.  I would STRONGLY recommend calling them at this point, if that is an option, as it will be much easier to try to restore your array.  If not, then just know that, while it is likely we can get it working again, there are no guarantees that the data is gone forever.  Before starting:

• Do you have a backup?
• How many drives total?
• How large are the disks?
• What was the total disk space Windows was using for all partitions?

hi flash, thanks for the advice & quick response. i tried to called the US support but couldn't make it through. to be honest, i dont have much experience with RAID so i would appreciate if u could guide me through to get my machine up & running again. to answer your question:

1. i dont have backup

2. 2 drives in total (C & D)

3. 120GB each

4. not sure this correct, it was saying the C drive is using 30GB

You have two physical drives or two drives listed in Windows (C: and D:)?  

RAID 5 must have a minimum of three physical disks.

120GB is an oddball size for SCSI ... are you sure that is the size of the physical disks? 

If you have a C: that is 30GB, do you remember the size of D:?

Sorry for the questions, but they all matter (another reason why it is easier on the phone with Support).  We can proceed without knowing this, but it is easier to go in knowing what we're working with.

4 physical drives with 2gb on each disk

dang, i i blew it again, 72gb actually, seagate 373207lc

sorry, i should be more detail. those are drives listed in windows. the system have 4 physical drives with 2gb on each disk. forget about what i said about the 30gb windows.

No problem :)

K ... power on and wait/look for the PERC to load - you will see a prompt to hit CTRL-M to enter the configuration utility.  Enter it, then go to Configure, View/Add Configuration.  If prompted between Disk View and NVRAM View, select Disk View, and type exactly what you see - don't do anything else right now.

there are 2 columns.

in RAID Ch-0: ID0 FAIL A00-00, ID1 Ready, ID2 FAIL A00-02, ID3 ONLIN A00-01, ID4 empty, ID5 empty, ID6 PROC

in SCSI Ch-1: all empty

i pulled #1 as 0:1 & #2 as 0:2

You said you got a message that disk 0:0 failed, then you pulled disk 0:1, and you said it booted to Windows ok?  I'm afraid this isn't possible.  A RAID 5 with four disks MUST have any three of the drives ONLINE at any given time in order for ANY of the data (including boot files) to be accessible at all.  With disk 0:0 failed, as soon as the controller detected that disk 0:1 was also not present/offline/failed, it could not have continued, as it would not have been able to recognize any of the data on the RAID 5 - the RAID 5 itself would be considered offline/dead by the controller, and it would have stopped right there with an error message.

So, my question is, how sure are you about exactly what happened (which drive was reportedly failed and which drive(s) was pulled)?

How critical is the data on this array (as you don't have a backup)?

Drive slots are labeled across the top, left to right:  1,3,5; and across the bottom, left to right:  0,2,4

yup, first i pulled out the top left, then middle bottom

So, when you said you pulled disk "#1", you actually mean the first drive, in this case 0:0, right?  or did you pull disk 0:1?  Check the ID on the chassis itself of the drive you pulled FIRST, then SECOND.

So here is the deal ... with the drives showing as follows:

ID 0:0 FAILED
ID 0:1 READY
ID 0:2 FAILED
ID 0:3 ONLINE

If 0:0 was failed before all this happened, then drive ID's 0:1, 0:2, and 0:3 must all be ONLINE (or made to be ONLINE) in order to attempt to access the data on the array.  With multiple offline drives, you must force online the appropriate drives (0:1 and 0:2 in this case); however, drives in a READY state cannot be forced online.  Meaning that, given all the above, the only way to do this is to "retag" the array, which re-assigns RAID settings/configuration to the drive(s) without initializing the array.

So, from the sounds of it, if this is unsuccessful in restoring your OS data, other than being somewhat inconvenient/time-consuming/costly to reinstall/rebuild the server, you will not have lost any critical data.  Is that correct?  The reason I ask is because if you MUST recover the data and would be considering professional data recovery options, then you need to stop here and not do any more to the drives before calling them.

I would strongly recommend calling Dell to walk through this procedure with Support on the line.  Very seldom (usually during the busy times of day, like the start of business day, and when there is the smallest staff, like evening/overnight) is the wait ever more than 2-3 minutes.  800-822-8965

Again, I would be happy to offer my assistance if you choose not to call Support.  Let me know what you want to do ... I'll be back on in the morning around 9AM MDT. 

Good luck.

i' afraid that is what i have. i still got the snapshot of the PERC error & i am sure i pulled 0:1 first then 0:2.

data that been stored in this array are the data for SQL. the data then be auto transferred to another server & this server will automatically delete the trasnferred data to maintain the storage capacity. so in terms of data backup, i think i do have it. but as for the OS, installed softwares etc... i pretty much sure i dont have it.