Logical Drive Offline, Disk Array Degraded

  • 905 Views
  • Last Post 22 November 2017
Thomas Klaey posted this 21 November 2017

Hi all,

i've a big issue with my R6.

The systems runs 24/7 and mostly without any issue.

Yesterday my 2 year old boy was able to reach out for the R6 and took out 2 disks :-(

This disks have then be marked as Offline of course, but also after a shout down, insert the disks and power up the disks still remained offline. Due to help i found here i was able to get the disks back runing but now i still have this 2 problems:

- Logical drives tells me it's offline
- Disk Arry tells me Degraded, PFA

My questions now:

- how can i bring back the Logical drive online?
- how can i bring back the Disk Array? (Rebuild of the Disk Array doesnt work, get an error message the disk are offline)

Is there any chance to bring back the system working or is everything lost?

Order By: Standard | Latest | Votes
Thomas Klaey posted this 21 November 2017

Find here some pictures of the system:

 

 

P B posted this 21 November 2017

From your description above, it sounds like you have (2) drives offline.

There are (2) methods:

  • Forcing the drives online.
  • Recreate the array (assuming all the required disks are OK - You should contact Technical Support for this method).

1. Please use promiseutil

HostName:~ user$ promiseutil

2. Use the phydrv command to check the physical drive status. Locate the disks that have a Dead or PFA operational status.

cliib> phydrv
===============================================================================
PdId Model        Type      Capacity  Location      OpStatus  ConfigStatus     
===============================================================================
1    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot1   OK        Array0 No.0      
2    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot2   OK        Array0 No.1      
3    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot3   OK        Array0 No.2      
4    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot4   OK        Array0 No.3 
5    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot5   Dead      Array0 No.4
6    Hitachi HUA7 SATA HDD  2TB       Encl1 Slot4   PFA       Array0 No.5

Can you post the output of your disks on this forum posting?

3. Check which disk is Dead and which disk is PFA.

4. You can use use the following command to force a drive online:
(x) = PdID (Physical Drive ID)

phydrv -a online -p(x)

4. If you see a drive with PFA, that states Predictive Failure Analysis (could be bad sectors/errors on the drive) and you should replace that drive with a new one (look at the compatability list to ensure you purchase a qualified/tested drive) - then rebuild the missing sequence. 

5. To bring the Logical Drive back to Critical status, you need to have 5 drives online.

  • With all 6 drives present, the Operational Status will be OK and not Critical. 
  • You can only lose one drive in a RAID5 LUN; hence, 5 of the drives need to be online in order to access the data since you created a RAID5 Logical Drive out of 6 disks.

6. If you do get your volume back online, I suggest running a filesystem via Disk Utility > First Aid just in case.

There are better tools out there:

This question was also answered in the following forum posting:
https://forum.promise.com/thread/r4-array-bad/

For further technical support, please contact support by opening a case @ https://support.promise.com

Thomas Klaey posted this 21 November 2017

Dear P B,

 

thanks for your answer.

I had 2 dead drives before which i replaced and are unconfigured actually.

phydrv reuslt is this:

 

cliib> phydrv

===============================================================================

PdId Model        Type      Capacity  Location      OpStatus  ConfigStatus     

===============================================================================

1    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot1   OK        Array0 Seq. No.0 

2    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot2   PFA       Array0 Seq. No.1 

3    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot3   OK        Unconfigured     

4    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot4   OK        Array0 Seq. No.3 

5    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot5   OK        Unconfigured     

6    Hitachi HDS7 SATA HDD  2TB       Encl1 Slot6   OK        Array0 Seq. No.5 

 

Which means drive 1 is PFA and i need to replace too.

So actually i have at least 5 online drives, but still cannot rebuild.

Does that mean i have to rebuild everything and will lose all the content?

Richard Oettinger posted this 21 November 2017

What does the "array -v" command show?

Thomas Klaey posted this 21 November 2017

array -v result:

DaId: 0

OperationalStatus: Degraded, PFA

Alias: DATA

PhysicalCapacity: 8TB                  ConfigurableCapacity: 12TB

FreeCapacity: 0Byte                    MaxContiguousCapacity: 0Byte

AvailableRAIDLevels: 0 5 6 10 50 1E

PDM: Enabled                           MediaPatrol: Enabled

NumberOfPhysicalDrives: 6              NumberOfLogicalDrives: 1

NumberOfDedicatedSpares: 0

 

Physical Drives in the Array:  

===============================================================================

SeqNo PdId CfgCapacity FreeCapacity OpStatus                                   

===============================================================================

0     1    2TB         401.41KB     OK                                         

1     2    2TB         401.41KB     OK, PFA                                    

2     3    2TB         401.41KB     Dead, Not usable                           

3     4    2TB         401.41KB     OK                                         

4     5    2TB         401.41KB     Dead, Not usable                           

5     6    2TB         401.41KB     OK                                         

 

Logical Drives in the Array:  

===============================================================================

LdId Alias          RAIDLevel Capacity  OpStatus                               

===============================================================================

0    DATA           RAID5     10TB      Offline                                

 

Available Spares to the Array:  

===============================================================================

Id  OpStatus  PdId CfgCapacity Revertible Type      DedicatedToArray          

===============================================================================

No spare drive available in the array

 

Richard Oettinger posted this 21 November 2017

One more cammand: logdrv -v

Thomas Klaey posted this 21 November 2017

logdrv - v result:

 

LdId: 0

ArrayId: 0                             SYNCed: Yes

OperationalStatus: Offline

Alias: DATA

SerialNo: 000000000000000000000000275842FD04049884

WWN: 229f-0001-55ae-c9b1               PreferredCtrlId: N/A

RAIDLevel: RAID5                       StripeSize: 128KB

Capacity: 10TB                         PhysicalCapacity: 12TB

ReadPolicy: ReadAhead                  WritePolicy: WriteBack

CurrentWritePolicy: WriteBack

NumOfUsedPD: 6                         NumOfAxles: 1

SectorSize: 512Bytes                   RAID5&6Algorithm: right asymmetric (4)

TolerableNumOfDeadDrivesPerAxle: 1     ParityPace: N/A

CodecScheme: N/A

 

Richard Oettinger posted this 21 November 2017

It appears that your Disk Array and Logical Drive were using all six of the drives:

NumberOfPhysicalDrives: 6

NumOfUsedPD: 6

The Logical Drive is a RAID5 - which can tolerate the loss of ONE drive. When you replaced drives 3 and 5 you lost TWO drives from the array, so you have effectively lost your data.

Do you still have those two drives? With them there is a chance the array can be rebuilt, and you should probavly open a case with Promise to attempt the recovery...

 

Thomas Klaey posted this 21 November 2017

i have the two drives, but they appear as unconfigured now, not dead... still a chance of getting the array back?

 

Richard Oettinger posted this 21 November 2017

Since we cant know which slot they were in, there are ways to test with one at a time to possibly rebuild the array.

But you are going to have to open a case with Promise to attempt the recovery.

Thomas Klaey posted this 21 November 2017

I know in which slot each of them were... the problem seems to be that the logical drive is offline. that is why i get an error trying to rebuild.

Richard Oettinger posted this 22 November 2017

You can not rebuild a RAID5 with two "dead" drives - a RAID5 can tolerate the loss of ONE drive only. With two dead drives any RAID5 would be offline.

Search
Categories
Close