]>
Commit | Line | Data |
---|---|---|
aa039b0f DM |
1 | Local ZFS Pool Backend |
2 | ---------------------- | |
fc3425bd | 3 | include::attributes.txt[] |
aa039b0f DM |
4 | |
5 | Storage pool type: `zfspool` | |
6 | ||
7 | This backend allows you to access local ZFS pools (or ZFS filesystems | |
8 | inside such pools). | |
9 | ||
10 | Configuration | |
11 | ~~~~~~~~~~~~~ | |
12 | ||
13 | The backend supports the common storage properties `content`, `nodes`, | |
14 | `disable`, and the following ZFS specific properties: | |
15 | ||
16 | pool:: | |
17 | ||
18 | Select the ZFS pool/filesystem. All allocations are done within that | |
19 | pool. | |
20 | ||
21 | blocksize:: | |
22 | ||
23 | Set ZFS blocksize parameter. | |
24 | ||
25 | sparse:: | |
26 | ||
27 | Use ZFS thin-provisioning. A sparse volume is a volume whose | |
28 | reservation is not equal to the volume size. | |
29 | ||
30 | .Configuration Example ('/etc/pve/storage.cfg') | |
31 | ---- | |
32 | zfspool: vmdata | |
33 | pool tank/vmdata | |
34 | content rootdir,images | |
35 | sparse | |
36 | ---- | |
37 | ||
38 | File naming conventions | |
39 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
40 | ||
41 | The backend uses the following naming scheme for VM images: | |
42 | ||
43 | vm-<VMID>-<NAME> // normal VM images | |
44 | base-<VMID>-<NAME> // template VM image (read-only) | |
45 | subvol-<VMID>-<NAME> // subvolumes (ZFS filesystem for containers) | |
46 | ||
47 | `<VMID>`:: | |
48 | ||
49 | This specifies the owner VM. | |
50 | ||
51 | `<NAME>`:: | |
52 | ||
53 | This scan be an arbitrary name (`ascii`) without white spaces. The | |
54 | backend uses `disk[N]` as default, where `[N]` is replaced by an | |
55 | integer to make the name unique. | |
56 | ||
57 | ||
58 | Storage Features | |
59 | ~~~~~~~~~~~~~~~~ | |
60 | ||
61 | ZFS is probably the most advanced storage type regarding snapshot and | |
62 | cloning. The backend uses ZFS datasets for both VM images (format | |
63 | `raw`) and container data (format `subvol`). ZFS properties are | |
64 | inherited from the parent dataset, so you can simply set defaults | |
65 | on the parent dataset. | |
66 | ||
67 | .Storage features for backend `zfs` | |
68 | [width="100%",cols="m,m,3*d",options="header"] | |
69 | |============================================================================== | |
70 | |Content types |Image formats |Shared |Snapshots |Clones | |
71 | |images rootdir |raw subvol |no |yes |yes | |
72 | |============================================================================== | |
73 | ||
74 | Examples | |
75 | ~~~~~~~~ | |
76 | ||
77 | It is recommended to create and extra ZFS filesystem to store your VM images: | |
78 | ||
79 | # zfs create tank/vmdata | |
80 | ||
81 | To enable compression on that newly allocated filesystem: | |
82 | ||
83 | # zfs set compression=on tank/vmdata | |
84 | ||
85 | You can get a list of available ZFS filesystems with: | |
86 | ||
87 | # pvesm zfsscan | |
deb4673f DM |
88 | |
89 | ifdef::wiki[] | |
90 | ||
91 | See Also | |
92 | ~~~~~~~~ | |
93 | ||
f532afb7 | 94 | * link:/wiki/Storage[Storage] |
deb4673f DM |
95 | |
96 | endif::wiki[] |