MSA2050 degraded drive

[javier12312]

Hi,

I've a MSA2050 with a degraded disk not failed. I've 1 disk configured as a global spare. What is the right procedure to replace that disk? Is there an option to failed the disk?

[JonPaul]

@javier12312 
A degraded drive does not need immediate replacement but if the system is stating it is degraded then it has seen issues with the drive and suspects that it could soon fail.  When the issues become more serious the system will FAIL the drive and rebuild to a spare drive.

If you do want to pre-emptively replace the degraded drive.
First check to see if a disk-group scrub has completed on that disk-group recently.  This will let you know that an end to end read of the disk-group was successful and other drives are not experiencing issues.

  A. Make sure that you have a good backup -- this is just in case things go catastrophic during the rebuild
  B. Make sure you have an appropriate replacement drive
  C. Make sure the disk-group is not degraded for any other reason than this disk 
  D. Validate that your spare drive is in a slot that is acceptable to your preferrence -- MSA traditional RAID sets will rebuild to the SPARE slot and will not return to the original slot. If you want to keep your disk-group physically in the same slots with traditional RAID levels then unconfigure and remove the spare drive(s) and put the spare drive in the slot of the degraded drive (make sure to give >30 seconds from drive removal to new drive insertion) if you have dynamic spares enabled the drive should be pulled in to reconstruct in a couple minutes, if not then add the drive as a SPARE.
  E. Set the Drive identify LED on/enabled -- Gen5 and lower is a BLUE LED,  Gen6 is the AMBER LED on for 3 seconds off for 1 second
      If something doesn't look right with the placement of the LED on drive do a thorough review of where the drive is and what drive you think should be removed.
  F.  Remove the drive -- this is not recommended as it is intentionally degrading the fault tolerance of your disk-group
     If you didn't do a good job in step C you, and all your customers, will know right away as data access halts.
  G. Wait a few minutes and see if reconstruction starts 
Depending on a number of factors reconstruction can take a long time.  It would be best to reconstruct when there is low I/O to the system.  During the reconstruction your disk-group is degraded or Critical and your data is at risk.  Once you remove the drive there is no going back, a reconstruction must complete for your data to be fault tolerant again.  

If you have an MSA Gen6 system and are using the MSA-DP+ RAID level then reconstruction will happen internal to the disk-group.  A drive placed into the removed drive slot will automatically be pulled into the disk-group and rebalance will occur.
Information about this and other topics is available in the MSA 6th Gen Virtual Storage Technical Guide:  https://www.hpe.com/psnow/doc/a00103247enw

[javier12312]

Hi,

@JonPaulthank you, very clear your explanation. I'll monitor the disk and I'll replaced when the disk fail.

Do you recommend update MSA firmware with the degraded disk?

Regards

[JonPaul]

@javier12312 
it is a best practice to keep your system and disk firmware up to date. There are checks within the system and in the Array Firmware Smart Component which will notify if an upgrade is not recommended at that time and what should be reviewed to successfully update.

For other best practice recommendation, please take advantage of the MSA HealthCheck:  https://www.hpe.com/storage/MSAHealthCheck