PURPOSE

Provide an understanding of the Data Domain (DD) Disk reconstruction process.

This article explains how disk reconstruction works on Data Domain systems and how to monitor reconstruction progress following a disk failure. Understanding this process helps ensure continued data protection and allows you to identify when additional actions are required.

Related Information
If you received a disk failure alert (EVT-STORAGE-00002) and are waiting for a replacement disk, see Data Domain: Failed, Absent, Unknown, or Powered-off Disk.
This article provides a disk replacement checklist, including steps to verify reconstruction activity and maintain system stability.

Understanding Data Domain Disk Reconstruction:

Data Domain systems use RAID 6 to protect data, even in the event of multiple disk failures. Monitoring disk reconstruction status is critical to ensure that the system continues to operate within protected limits.

• RAID 6 Protection: Supports multiple simultaneous disk failures without data loss (RAID 6 has double parity protection per disk group).

• Reconstruction Rules:

  • A maximum of two disk reconstructions can occur simultaneously (across two disk groups).
  • Exception: If two disks fail within the same Disk Group (DG), only one reconstruction can proceed at a time. The second will begin only after the first completes.

• Global Sparing Behavior:

  • Data Domain uses global spares across all external enclosures.
  • If all spares in one enclosure are consumed and not replaced, the system pulls a spare from another enclosure with the matching disk size.
  • This may result in a Disk Group spanning multiple enclosures. While it is ideal to keep RAID and spare disks within the same enclosure, this configuration does not impact performance.

• Reconfiguration Guidance: If global sparing has occurred, contact support to verify system health and assist with disk enclosure reconfiguration.

• Controller Disk Behavior: Head Unit (Controller) disks do not reconstruct to external shelf spare disks.

• Best Practice: Replace failed disks as soon as possible to maintain optimal protection and system stability.

Monitoring Disk Reconstruction Status on Data Domain 
To monitor disk reconstruction progress on a Data Domain system, use the following DD-CLI command:

# disk show state
OR
# storage show all

Command Output Includes:

• Current disk state
• Reconstruction activity
• Disk group undergoing reconstruction
• Percent completion of reconstruction

Special Case: If a second disk failure occurs in the same disk group (for example, dg1 or dg2), the output indicates a reconstruction pending status. This means that the second reconstruction does not begin until the first has completed.

sysadmin@hostname# disk show state
Enclosure   Disk
            1  2  3  4  5  6  7  8  9  10 11 12 13 14 15
---------   ---------------------------------------------
1           s  .  .  .
2           .  .  R  .  .  .  .  .  .  .  .  .  .  .  F
3           F  .  .  .  .  .  .  .  .  .  .  .  .  .  R
4           .  s  .  F  .  .  .  .  .  .  .  .  .  .  .
---------   ---------------------------------------------

Legend       State                          Count
------       ----------------------------   -----
.            In Use Disks                   44
s            Spare Disks                    1
R            Spare (reconstructing) Disks   2
F            Failed Disks                   3
------       ----------------------------   -----
Total 49 disks

Disk      Disks                        Progress   Remaining
Group   Reconstructing                 (%)        (minutes)
-----       --------------            --------    ---------
dg1           2.3                      0             238
dg2           3.15                    99             10
dg3        (Pending)
-----       --------------            --------    ---------