[Last Call] Learn how to a build a cloud-first strategyRegister Now

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 2646
  • Last Modified:

Import foreign configuration to PERC 5/E on MD1000

Hi Experts,

 my virtual disks disappear and all physical disks become in foreign state , i tried to import foreign configuration through openmanage or Ctrl+R utility but it failed with error code  "unknown"

attached file for PERC 5/E Adapter  .
lsi-0827.log
0
Mohamed Hammad
Asked:
Mohamed Hammad
  • 8
  • 5
  • 3
1 Solution
 
Davis McCarnOwnerCommented:
Your log says drive 1 has failed, and, possibly several more as well.  They are reporting as not ready or busy which is not good.
0
 
Mohamed HammadAuthor Commented:
These drivers are not part of the virtual disks , we have just one drive has failure predicted and I think even if it was failed the array should work .
0
 
Davis McCarnOwnerCommented:
You're asking the controller to rebuild the array configuration from data found on the drives.  With one drive reporting not ready and (I think) two more reporting busy, the controller can't complete that operation.
What made you think the controller was bad in the first place?
0
Get quick recovery of individual SharePoint items

Free tool – Veeam Explorer for Microsoft SharePoint, enables fast, easy restores of SharePoint sites, documents, libraries and lists — all with no agents to manage and no additional licenses to buy.

 
Mohamed HammadAuthor Commented:
i don't say that the controller is bad , anyway i tried to change it and the same result , all disks are foreign and can't be imported  . do you think we can return it back
0
 
Davis McCarnOwnerCommented:
Either use the BIOS utility or Open Manage to check the state of every drive in the array, first!
Read the BIOS Configuration section of this:
http://web.mit.edu/cron/documentation/dell-server-admin/en/Perc5i_5e/chapterh.htm#wp1066465
0
 
Jim_NimCommented:
There is actually a "predictive failure" on both slot 1 and 10. This is not equal to an actual failure though, and these two drives might still be usable for now.
If all drives are showing as "foreign" as you've said, that means that none of the drives are in a failed state, and all are essentially "ready", with configuration data detected on all of them.

If your import is failing when all the drives are present, your options are limited. You're already on the latest firmware release available, so this may not be due to a problem with the controller.

2 options I would recommend considering:
If the data is extremely important (too costly to risk potentially damaging), reach out to a data recovery service (e.g. Drive Savers or Kroll OnTrack)
If data recovery services aren't an option, you can try a "retag" operation - essentially recreating the RAID configuration on top of the existing config

RAID retagging process:
Clear the foreign config (once this is done, it cannot be undone!)
Check to ensure that all drives show as being in a ready/available state
Recreate your configuration from the PERC BIOS, ensuring to choose the same RAID type (most other settings should typically be left with default settings). Ensure that you do NOT start an initialization when prompted, as this will overwrite all of the data on the virtual disk.
Boot to the OS and check to see if the file system is intact!
Get those 2 predFail drives replaced as soon as possible!
0
 
Mohamed HammadAuthor Commented:
I got the status , all physical disk at open manage or Bios utility are foreign , and why trying to import  from openmange it said "the operation can't be performed" and from bios it said "Error : unknown"
0
 
Mohamed HammadAuthor Commented:
Jim_Nim ,

that's exactly what I reach at the end , the most risky step with the advanced option as I'm not sure if it was left with default or not , at this point exactly , do you know how to get the raid configuration from logs ?

I already have the span and drivers orders still need to be sure of advanced settings
0
 
Jim_NimCommented:
I would recommend that you try leaving the advanced settings at defaults, create the config, and test to see if data is visible. So long as you don't initialize, you can retry creating the config in different ways to see which is successful.

The controller logs typically never provide data that can be used to retrieve the configuration. However, if you ever called Dell support in the past and gathered a DSET report, we may have one saved in our database under your array or server's service tag that would have the configuration data saved.

I know it's probably been a long time but... do you remember where you created the original configuration at? If you configured things from within the GUI (OMSA) then you almost certainly used default settings.
0
 
Mohamed HammadAuthor Commented:
unfortunately I think we didn't provide the DEST report at Dell but if you can made me a favor and check for at at your database I will appreciate it , I sent your the service tag for our MD .

also are you confirming  that we can try with retagging in different ways as we didn't initialize ?

I was talking to Dell support and he confirmed that if we did it with a wrong way we will destroy the data .
0
 
Mohamed HammadAuthor Commented:
Jim ,

 Many thanks for your recommendation , I follow your steps and my data is back again .

 But , background initialization is started and i cancel but it restarted again after 2 or 3 minutes ,I'm afraid it erase my data or get the virtual raid foreign or failed again , what do you think ?

 Please advice.....................find attached the logs , I've exported it after retagging
lsi-0903.log
0
 
Jim_NimCommented:
If you are absolutely certain that "advanced > initialize" was not selected/checked when you retagged, then the "background initialization" reported is actually a consistency check, and the "background initialization" is a non-destructive one.

I'll try to verify this in the lab, though I only have access to a PERC 6.


Is your data still accessible now after a couple hours have passed?
0
 
Mohamed HammadAuthor Commented:
welcome back Jim , where you were all this time , we missed your guidance.

Yes my data is accessible  until now however I'm cancel background initialization each time it appear .
my physical disks after retagging were optimal but after a while it show me  the 01 physical disk is failure predicted .

do you think if i let background initialization finished , my virtual disk will return to foreign state again ?
0
 
Jim_NimCommented:
I highly doubt that will happen. If a background init was overwriting data, you would lose your partition data pretty quickly, which would lead to loss of access to the file system at the next disk rescan or reboot.

You actually have a prem/fail on both slot 1 and 10. Those two drives need to be replaced when possible.
0
 
Mohamed HammadAuthor Commented:
I'm very confused can't you assure me what background initialization do isn't it overwriting data or not you said it's not data destructive .

it's now 2 % and data still accessible  , should i keep server up without restart as u said restart may make the data inaccessible .
0
 
Jim_NimCommented:
I suspect that the PERC is over-using the term "background initialization" to mean more than just one thing, which makes it very unclear what it's doing now.  Hopefully the background init that it reports now is little more than just a consistency check.  

However, if you're unsure, performing a reboot might be a better idea than waiting it out...  Would you rather lose access to your data now, or overwrite all of it after having a few extra hours of access to it?
0

Featured Post

Free Tool: Port Scanner

Check which ports are open to the outside world. Helps make sure that your firewall rules are working as intended.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

  • 8
  • 5
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now