Start a Conversation

Unsolved

This post is more than 5 years old

DG

1534

September 8th, 2012 10:00

Two disks failed from same RAID group

We have identified two separate disks that have failed simultaneously. 1B-C3 and 16C-C5 in symmetrix.
Unfortunately they are both in the same raid group. Raid-5 (7+1).
We are currently unable to initiate a replacement of either Drive, because of potential data loss on the other drive.

Total 5 devices affected. These are assigned to different servers.

Any procedure to replace the disks without data loss. I really appreciate your help.


2 Intern

 • 

5.7K Posts

September 8th, 2012 10:00

An SR should have been created for this and since it's in the same raid group I think it's a prio 1 service request (sev 1). I would EMC have a saying in this but I thing yourmdata is toast and estoring it from a backup might be needed.

9 Legend

 • 

20.4K Posts

September 8th, 2012 21:00

if it's not hard crash, support might be able to spin it back up.

1.4K Posts

September 8th, 2012 22:00

DFRG's seems to be a trending issue now a days.

I am doing that from last 4 days constantly till today! Even glen would agree with me on that! :/

September 15th, 2012 08:00

#symdisk list -failed not shows any failed disks. No hotspare also invoked. Here is the output of symdev show output.

Device : 04AC (m)

     {

     --------------------------------------------------------------

      Disk     DA       Hyper       Member    Spare       Disk

     DA :IT   Vol#   Num Cap(MB)  Num Status  Status  Grp#  Cap(MB)

     --------------------------------------------------------------

     15A:D1    578    57    1247    1 RW      NR         1   139814

     16A:D2    662    56    1247    7 RW      NR         1   139814

     01B:C3    128    58    1247    6 NR      RW         1   139814

     15B:C3    124    58    1247    8 RW      NR         1   139814

     02C:D2    626    56    1247    5 RW      NR         1   139814

     16C:C5    196    58    1247    4 NR      NR         1   139814

     01D:D2    688    56    1247    3 RW      NR         1   139814

     02D:C0     57    57    1247    2 RW      NR         1   139814

     }

The particular disks 01B:C3 , 16C:C5 are in NR and 01B:C3 is having spare status RW. Till now i didn't get any issues from platform team and application team.

Like the same  for all 5 affected devices.

Please let me know how to proceed.

2 Intern

 • 

448 Posts

September 17th, 2012 09:00

Have you contacted EMC yet?  I believe you can only have one spare applied against a raid group at a time by the automatic sparing.  It does appear that one drive spared out so while you have two failed drives only one is truly unavailable.  You should be able to have one of the disks replaced again I would contact EMC support and let them determine which one to replace first.

I would expect that if you replace the drive that is spared it will build from the spare.  If you replace the drive that is down and not spared it will have to build off parity calculations.

No Events found!

Top