NetApp - ONTAP disk ownership, aggregates and volumes as built
NetApp Solution · Config document · referenced from Storage design
noteThis is a listing of state, not a command set. The notes on disk assignment contain no
storage disk assign, storage aggregate create or volume create command, and none are invented here. The only commands in the notes are the ones that display the state; they head the blocks below.The storage configuration of one cluster as it was recorded: shelves, disk ownership by node and pool, aggregates with their plexes, volumes, and the policies on them. It is the result of the disk assignment and the aggregate layout, without the steps that led there.
| Item | Value |
|---|---|
| Shown here | Cluster DC1-A-XNAS001, datacenter A of site 1 |
| Also applies to | DC1-B-XNAS001; see the differences below |
| Sources | The as-built report produced with NetAppDocs, the disk layout spreadsheet of the design, and storage shelf show from the notes of September 2017 |
| Columns | Those of the sources, arranged under the ONTAP command that shows the object; not the exact output of the commands |
| ONTAP version at the time | 9.1P11 |
| State of the SVMs | Four data SVMs; the volumes of the two SVMs added in 2019 are not in the report |
The file
# --- storage shelf show ------------------------------------------------------ # Output of September 2017. Stack numbers are assigned by the cluster and # were different in later listings; shelf IDs are fixed. # Shelf ID Model Module Status Datacenter Disks 1.30 30 DS224-12 IOM12 Normal A 24 1.31 31 DS224-12 IOM12 Normal A 16 2.10 10 DS224-12 IOM12 Normal A 24 2.11 11 DS224-12 IOM12 Normal A 24 3.20 20 DS224-12 IOM12 Normal A 24 3.21 21 DS224-12 IOM12 Normal A 24 4.70 70 DS224-12 IOM12 Normal B 24 4.71 71 DS224-12 IOM12 Normal B 16 5.60 60 DS224-12 IOM12 Normal B 24 5.61 61 DS224-12 IOM12 Normal B 24 6.50 50 DS224-12 IOM12 Normal B 24 6.51 51 DS224-12 IOM12 Normal B 24 # --- storage disk show -ownership, summed up --------------------------------- # Disk type 1.2TB_SAS_10k, 1.09 TB usable. Pool0 is the datacenter of the # owning node, Pool1 the other one. # Owner Disks In aggregates Spare Pool0 (shelves) Pool1 (shelves) DC1-A-ANAS001 130 58 72 10 to 30 and 5 of 31 50 to 70 and 5 of 71 DC1-A-ANAS002 120 14 106 10 to 30 50 to 70 DC1-B-ANAS001 16 14 2 71 31 DC1-B-ANAS002 6 4 2 71 31 # --- storage aggregate show -------------------------------------------------- # All aggregates: online, mirrored, normal. # Aggregate Node RAID Disks RG size HA policy Root Size DC1_A_ANAS001_data1 DC1-A-ANAS001 raid_dp 54 16 sfo no 21.42 TB DC1_A_ANAS001_root DC1-A-ANAS001 raid4 4 8 cfo yes 953.80 GB DC1_A_ANAS002_data1 DC1-A-ANAS002 raid_dp 10 16 sfo no 2.79 TB DC1_A_ANAS002_root DC1-A-ANAS002 raid4 4 8 cfo yes 953.80 GB # --- storage aggregate plex show --------------------------------------------- # Plex names from the disk layout spreadsheet; disk counts are half of the # aggregate. # Aggregate Plex Pool Datacenter Disks DC1_A_ANAS001_data1 plex0 0 A 27 DC1_A_ANAS001_data1 plex1 1 B 27 DC1_A_ANAS001_root plex0 0 A 2 DC1_A_ANAS001_root plex4 1 B 2 DC1_A_ANAS002_data1 plex0 0 A 5 DC1_A_ANAS002_data1 plex1 1 B 5 DC1_A_ANAS002_root plex0 0 A 2 DC1_A_ANAS002_root plex4 1 B 2 # --- volume show ------------------------------------------------------------- # <ID> stands for the identifier MetroCluster puts into the names of its # metadata volumes. # Vserver Volume Aggregate Size Snapshot policy Style DC1-A-ANAS001 vol0 DC1_A_ANAS001_root 902.54 GB - - DC1-A-ANAS002 vol0 DC1_A_ANAS002_root 902.54 GB - - DC1-A-XNAS001 MDV_CRS_<ID>_A DC1_A_ANAS001_data1 10.00 GB none unix DC1-A-XNAS001 MDV_CRS_<ID>_B DC1_A_ANAS002_data1 10.00 GB none unix DC1-S-VCVSM001 DC1_S_VCVSM001_root DC1_A_ANAS002_data1 1.00 GB default unix DC1-S-VCVSM001 DC1_S_VCVSM001_data DC1_A_ANAS002_data1 100.00 GB default unix DC1-S-VCVSM002 DC1_S_VCVSM002_root DC1_A_ANAS001_data1 1.00 GB default unix DC1-S-VCVSM002 DC1_S_VCVSM002_data DC1_A_ANAS001_data1 10.10 TB none unix DC1-S-VCVSM003 DC1_S_VCVSM003_root DC1_A_ANAS001_data1 1.00 GB default ntfs DC1-S-VCVSM003 DC1_S_VCVSM003_data DC1_A_ANAS001_data1 100.00 GB default ntfs # --- volume show -fields space-guarantee, autosize, efficiency --------------- # The same on every volume above. space guarantee = volume deduplication = not enabled compression = not enabled autosize = disabled try first when full = volume_grow snapshot autodelete = disabled autodelete trigger = volume target free space = 20% # --- volume snapshot policy show --------------------------------------------- # Policy Enabled Schedules: copies kept default true hourly: 6, daily: 2, weekly: 2 default-1weekly true hourly: 6, daily: 2, weekly: 1 none false - # --- job schedule show, the schedules the policies use ----------------------- hourly = every hour at minute 5 daily = every day at 0:10 weekly = Sunday at 0:15 # --- volume efficiency policy show ------------------------------------------- # Present in every data SVM, used by no volume. # Policy Enabled Schedule QoS policy default true daily best_effort inline-only - - -
What changes on cluster B
| Object | DC1-A-XNAS001 | DC1-B-XNAS001 |
|---|---|---|
| Shelf names | As above | The same shelves as 1.50, 1.51, 2.60, 2.61, 3.70, 3.71, 4.10, 4.11, 5.30, 5.31, 8.20, 8.21 |
| Data aggregates | One on each node, 54 and 10 disks | One only: DC1_B_ANAS001_data1, raid_dp, 10 disks, 2.79 TB |
| Root aggregates | DC1_A_ANAS001_root, DC1_A_ANAS002_root | DC1_B_ANAS001_root, DC1_B_ANAS002_root, same size and RAID type |
| Plexes | plex0 in datacenter A; mirror plex1 or plex4 in B | plex0 in datacenter B; mirror plex1 for the data aggregate and plex4 for the root aggregates in A |
MDV_CRS volumes | One on each data aggregate | Both on DC1_B_ANAS001_data1 |
| SVM volumes | Root and data volumes of three SVMs | DC1_S_VCVSM004_root only, 1 GB, security style ntfs, snapshot policy default |
| Snapshot policies | default, default-1weekly, none | The same three; on each cluster the policies of the mirrored SVMs of the partner also appear, with the suffix -DR |
What changed later
| Object | As-built report, 2018 | Final design, 2019 |
|---|---|---|
DC1_A_ANAS001_data1 | 54 disks | 64 disks |
| Volumes on it | Those above | Also DC1_S_VCVSM002_data2, and root and data volumes of DC1-S-VCVSM005 and DC1-S-VCVSM006 |
Site 2, DC2_A_ANAS001_data1 | Not built yet | 38 disks |
| Aggregate names | DC1_<datacenter>_<node>_data1 and _root | Unchanged. In September 2017 they were still data_DC1_A_ANAS001, root_DC1_A_ANAS001 and so on |
The commands with which the state was looked at are in Storage design.
Checked against ONTAP 9.19.1
| As built | Today |
|---|---|
| Pool 0 local, pool 1 remote | Unchanged for MetroCluster FC. Disks are assigned to a pool with storage disk assign, which keeps its -pool and -owner parameters, or in maintenance mode with disk assign -shelf and -p 0 or -p 1. NetApp states that new MetroCluster systems are shipped with the disk assignment done |
| Mirrored aggregates | storage aggregate create with -mirror true and storage aggregate mirror are unchanged |
| Whole-disk root aggregates | Root-data partitioning is still listed as not supported for fabric-attached MetroCluster; it is supported for MetroCluster IP from ONTAP 9.4 |
storage disk show -ownership, storage disk option show, storage aggregate show -disk | All still in the command reference |
| DS224C shelves with IOM12 modules | Not among the shelves whose support was discontinued |
| FAS8200 | Support discontinued from ONTAP 9.17.1; the last release for it is 9.16.1 |
The first row explains part of what the notes lack. A MetroCluster arrives with its disks assigned, so there may never have been a full set of assignment commands to keep; what was done by hand here was the later rearrangement of the short shelves, and that is the part that is missing.