StorageGRID node reports zero or too low for replicated or erasure coding data
Applies to
NetApp StorageGRID 11.x
Issue
The issue is functionally cosmetic as it does not impact GRID function. It is when overhead is calculated during one of the following operations:
- Node creation during deployment.
- Node creation during expansion.
- Node recovery.
- Node reboot
- Upgrade
After, the values for one or more Object Store volumes (RangeDBs) show zero usage or a too-low amount.
NOTE: This can go unnoticed until a sizable amount of data has been ingested.
This can be identified by one or more of the following examples:
- StorageGRID node LDR overview for individual volumes shows inconsistent values for "Replicated Data" or "EC Data".
- Space usage dropped after the hardware replacement
- One node showing a different used %:
- One site shows a higher usage than the other when using a copy per site ILM rule:
- The usage is different from the size reported by ONTAP within a Fabric Pool configuration.