HPE 3PAR StoreServ Storage

Spare chunklet and its behavior under disk failure

Go to solution
Occasional Advisor

Spare chunklet and its behavior under disk failure


First, we have a 3PAR configuration of 2 CPGs with AO:
1. RAID 6, with 14 data and 2 parity for 16 FC 10K SAS HDDs
2. RAID 6 with 4 data and 2 parity for 6 SSDs
- Default CPG: FC (moves to SSDs nightly with scheduled AO)
- Total raw capacity: 29TB
- Allocated space: >85%. (about 14% of allocated space is usused spare chunklets)

As seen from our 3PAR space consumption, about 14% of allocated space is consumed by the system as unused spare chunklets.

Our questions:

1. In case there are not enough disks, will the system still keep building up the spare chunklets?
2. Say, if one SSD has failed and there is not enough space in the SSD tier, will data be relocated to the FC tier automatically?
3. What is the maximum number of FC and SSD disks we can afford to fail at the same time?





Re: Spare chunklet and its behavior under disk failure

Hello Darienn,

1. No. The number of spare chunklets are fixed as per your SparingAlgorithm settings. The current setting can be seen by running the command 'showsys -param'. The available values are as below

Default (roughly 2.5% with minimums), Minimal (roughly 2.5% without minimums), Maximal (one disk's worth in every cage), and Custom (not managed automatically by the system).

If you need to alter this, use the command "setsys SparingAlgorithm <value>" and then "admithw". Normally you do not need to change the default settings.

2. Yes. If no spare chunklets available in SSD drives, the same will be allocated from FC drives if required

3. Since this array has RAID 6, it can withstand two drives failures in the same RAID set. If the question is related to spare chunklets, the array will use any free chunklets (if no spare chunklets available when a drive fails).


I am an HPE Employee
Accept or Kudo