Sun Microsystems, Inc.  Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-72-1006824.1
Update Date:2010-05-26
Keywords:

Solution Type  Problem Resolution Sure

Solution  1006824.1 :   Work around to maintain RAID configuration after upgrading LSI firmware from 1.04.00.00.  


Related Items
  • Sun Fire X4200 Server
  •  
Related Categories
  • GCS>Sun Microsystems>Servers>x64 Servers
  •  

PreviouslyPublishedAs
209481


Symptoms
The raid array is degraded and the LSI configuration utility does not allow the disks to be resynchronized.

The menu item for resynchronization is grayed out.
This can also happen if the motherboard is swapped and the replaced board had LSI firmware 1.04.00.00.



Resolution
Rebuild the RAID1 array keeping existing data on the primary drive via the LSI Logic Config Utility menu options.

Eventually we expect a utility from LSI logic to correct this issue.



Relief/Workaround

If you need to preserve existing data, then please contact SUN before you proceed further.

Control-C when prompted with the LSI bios to go into the LSI Logic Config Utility.

  • Select SAS1064
  • On this screen, make careful note of which drive is currently active and which one is failed. You will need this when you recreate the RAID volume.
  • Select Raid Properties
  • Select Manage Array
  • Choose Delete Array
  • Hit "Y" to "Delete array and exit to Adapter Properties"
  • Wait for the deletion to complete.
  • Choose Raid Properties
  • Choose Create RAID1 Volume

IMPORTANT! Move to the line of the disk that was previously the active one. This will be the drive you start your volume with.

  • Move the cursor over to the "RAID Disk" column (it will say "[NO]" for its value).
  • Hit the space bar to select it.
  • Choose "M" to "Keep existing data, migrate to an IM array. Synchronization of disk will occur." This is the option that tells the LSI controller to keep your data.
  • The Drive Status for this drive will now show "Primary" and the "RAID Disk" status will be "YES".
  • Move the cursor over to the "RAID DISK" column for the drive that was previously failed. It will say "[NO]" for its value).
  • Hit the space bar to select it.
  • If this drive did have data on it in the past, you may get a warning saying that DATA was found on the selected disk. Hit Escape to continue.
  • The Drive Status for this drive will now show "Secondary" and the "RAID Disk" status will be "YES".
  • Hit "C" to create the array.
  • Select "Save changes then exit this menu".
  • Choose Raid Properties.

You can see the status of the sync next to the Status field. It may take a few minutes for the status percentage to move beyond 0%.
The drive that was previously failed should show a status of "Not Syncd" at this point until it syncs up.
When the array is finished sync'ing, the status will be "Optimal"
Once syncing has begun the system can be rebooted.
Exit all the way out and choose the option to exit and reboot.



Additional Information
This is related to bug 6499176.

Synopsis: LSI RAID configuration degraded on moving disks to Firmware 1.10.00.00 from 1.04.00.00

Screen shot of degraded status:

* LSI Logic Config Utility          v6.06.06.00 (2006.03.22)                   *
* View Array -- SAS1064                                                        *
*      Array                      1 of 1                                       *
*      Identifier                 LSILOGICLogical Volume  3000                 *
*      Type                       RAID1                                        *
*      Scan Order                 0                                            *
*      Size(MB)                   69618                                        *
*      Status                     Degraded                                     *
*                                                                              *
*      Manage Array                                                            *
*                                                                              *
* Slot  Device Identifier             RAID   Hot    Drive      Pred  Size      *
* Num                                 Disk   Spr    Status     Fail  (MB)      *
*   0   FUJITSU MAV2073RCSUN72G 0301  Yes    No     Primary    No     69618    *
*   1   SEAGATE ST973401LSUN72G 0556  Yes    No     Not Syncd  No     69618    *
*                                                                              *
*                                                                              *
*                                                                              *
*                                                                              *
*                                                                              *
*                                                                              *
* Esc = Exit Menu       F1/Shift+1 = Help                                      *
* Enter=Select Item  Alt+N=Next Array  C=Create an array                       *
*******************************************************************************

Related product documentation
For further information on known problems and workarounds refer to the Sun Fire X4100/X4200 Servers Product Notes, available on the Sun Fire X4100/X4200 Server Installation Documentation page.


Product
Sun Fire X4200 Server

Internal Comments
If they Customer has an '8e' partition type then they may not be able to re-mirror and preserve the data once the mirror has been deleted as '8e' is not a supported partition type at this time, see CR 6497076.
If they require to keep there data then please see CR 6499176 which contains details on this issue.
Contact TSC-SERVER-VSP who can look to supplying a utility to change Raid Volume Page 0 Settings field (see CR for details of utility)
Further details around CR
Re-sync not starting after upgrading FW. This issue is specific to 1.04.00 FW ONLY. This version of FW did not have support for the auto-configure bit in Raid Volume Page 0 Settings field. Volumes created under 1.04.00 FW had this bit set to 0. Because the volume settings are stored in meta-data, when FW is upgraded to a version that has support for auto-rebuild, the volume will have automatic re sync disabled.
RAID, x4100, x4200, synchronization, degraded
Previously Published As
88023
Change History
Date: 2010-05-26
User Name: james.j.carter@oracle.com
Action: Checked for currency, inc reference to latest product documentation

Attachments
This solution has no attachment
  Copyright © 2011 Sun Microsystems, Inc.  All rights reserved.
 Feedback