Physical and Logical Storage Resources
What this domain tests
This domain covers how disks become aggregates, how aggregates become volumes and LUNs, and how space is actually consumed.
Core rules
- Aggregates are built from RAID groups using RAID-DP (double parity) or RAID-TEC (triple parity) and draw from hot spare disks; ONTAP reserves spares and warns when the spare count falls under the low threshold.
- Advanced Drive Partitioning puts the root partition and a data partition on the same disk pair, so small systems do not waste whole disks on the root aggregate.
- WAFL writes consistency points and handles Snapshots, deduplication, compression, and compaction; Snapshots consume space from the volume and, once the Snapshot reserve is exhausted, can spill into free space.
- Volume guarantee can be
none(thin, overcommitted) orvolume/file(reserved). Autosize can grow or shrink a volume within a maximum. - FlexVol is the standard volume; FlexGroup is a scale-out volume with multiple constituents in one namespace; FlexClone is a writable point-in-time copy that shares blocks until changed.
Worked example
Provisioning a 20 TB FlexVol on a 15 TB aggregate succeeds when the guarantee is none, because ONTAP does not reserve the space. Writes later fail with ENOSPC once the aggregate fills. The fix is to check volume show -fields size,available,percent-used,guarantee and either raise the guarantee or move the volume to an aggregate with real free space.
Traps
Treating percent-used as only user data - it includes Snapshot copies. Assuming a FlexClone immediately doubles space usage. Forgetting that aggregate Snapshot reserve and volume Snapshot reserve are separate settings.