This post is more than 5 years old
1 Rookie
•
38 Posts
0
192984
February 5th, 2016 06:00
Defunct Disks Everywhere!!
Gen 3 PE 1950 with 4 drives in RAID 5. This server also has a 6GBS HBA that attaches to a storage enclouse with 14 drives. We started having issues not being able to connect to the SQL DB's that reside on the storage enclosure. So i looked in the event logs and found:
- There was an unrecoverable disk media error during the rebuild or recovery operation: Physical Disk 0:0:2 Controller 0, Connector 0
- There was an unrecoverable disk media error during the rebuild or recovery operation: Physical Disk 0:0:1 Controller 0, Connector 0
- Unexpected sense. SCSI sense data: Sense key: 3 Sense code: 11 Sense qualifier: 0: Physical Disk 0:0:2 Controller 0, Connector 0
- An error was detected on device \Device\Harddisk0\DR0 during a paging operation
- A block on the physical disk has been punctured by the controller: Physical Disk 0:0:1 Controller 0, Connector 0
- A block on the physical disk has been punctured by the controller: Physical Disk 0:0:3 Controller 0, Connector 0
- A block on the physical disk has been punctured by the controller: Physical Disk 0:0:1 Controller 0, Connector 0
So i started looking at the storage enclosure immediately. After days of testing each physical drive in the enclosure, i login to OMSA on this server and see the internal drives connected to the PERC6/i labeled:
- Physical Disk 0:0:0
- Physical Disk 0:0:1
- Physical Disk 0:0:2
- Physical Disk 0:0:3
So, it looks like we may have many drive issues including the enclosure (during the physical testing of 14 drives, 3 failed). So i am trying to decide the best way to resolve this issue. My thought, is start with the 4 internal drives, then move to the enclosure. Starting with the PERC6/i drives, it looks like there may be issues with 3 drives? If so, remove disk 0 and install another disk, use OMSA to add it to the array, let the array rebuild. Do this all the way to disk 3, yes?
theflash1932
11 Legend
•
16.3K Posts
0
February 6th, 2016 09:00
If the VD has a puncture, it is incomplete/corrupt and will remain so as long as the VD is configured - it is irreparably broken and cannot be repaired. The only way to "fix" a puncture is to delete its VD and create a new one. Keep the original drives if you need them, but only use them to pull files if needed (or, as recommended, get a file-level backup now) ... using them as the primary data source only puts your data at further risk. The disks CAN however be re-used ... if the drives are healthy, putting them in a NEW VD is fine.
carnold5
1 Rookie
•
38 Posts
0
February 5th, 2016 13:00
WOW!!! The VD IS the OS drive...... This server OS is server 2012 R2 core
This server has 7 hyper-v VM's.
This started right after the drives firmware was updated. Knowing this, does this change the process in any way?
carnold5
1 Rookie
•
38 Posts
0
February 6th, 2016 08:00
In your instructions, it says to delete the VD and then replace any failed drives or drives marked as predictive failure. I think i am going to replace drive 0 and drive 1 and make it its own array and replace drives 2 and 3 and make it its own array. I would like to keep the original drives without deleting the VD in case i need some data off the drives. If i do it this way, will i run the chance of the puncture coming back?
carnold5
1 Rookie
•
38 Posts
0
February 9th, 2016 17:00
Robert O, the first link about consistency checks seems dead. I'm also not sure what is meant by, "with multiple VD's, time them". What do you mean time them? So our servers have multiple VD's and 1 physical controller installed but has 2 connectors according to OMSA. So each of my commands would look like this:
omconfig storage vdisk action=checkconsistency controller=0 vdisk=0
omconfig storage vdisk action=checkconsistency controller=0 vdisk=1
or would they look like this:
omconfig storage vdisk action=checkconsistency controller=0 vdisk=0
omconfig storage vdisk action=checkconsistency controller=1 vdisk=1
carnold5
1 Rookie
•
38 Posts
0
February 10th, 2016 17:00
Robert O - you and theflash have been a huge help in this process!! Thank you both so much!
Is there a way that you know of to have an email sent with the results of the consistency check?
carnold5
1 Rookie
•
38 Posts
0
February 13th, 2016 06:00
Robert 0 - I meant is there a way to have the results of the consistency check emailed to me or do i misinterpret your email?
carnold5
1 Rookie
•
38 Posts
0
February 15th, 2016 18:00
Yes, the array was rebuilt according to your instructions found above. We also used different physical drives. Can you tell me where in the exported report i can find the consistency check?