]>
Commit | Line | Data |
---|---|---|
aa039b0f DM |
1 | [[chapter-storage]] |
2 | ifdef::manvolnum[] | |
7e2fdb3d DM |
3 | PVE(1) |
4 | ====== | |
38fd0958 | 5 | include::attributes.txt[] |
5f09af76 DM |
6 | :pve-toplevel: |
7 | ||
aa039b0f DM |
8 | NAME |
9 | ---- | |
10 | ||
11 | pvesm - Proxmox VE Storage Manager | |
12 | ||
13 | ||
49a5e11c | 14 | SYNOPSIS |
aa039b0f DM |
15 | -------- |
16 | ||
17 | include::pvesm.1-synopsis.adoc[] | |
18 | ||
19 | DESCRIPTION | |
20 | ----------- | |
21 | endif::manvolnum[] | |
22 | ||
23 | ifndef::manvolnum[] | |
24 | {pve} Storage | |
25 | ============= | |
38fd0958 | 26 | include::attributes.txt[] |
aa039b0f DM |
27 | endif::manvolnum[] |
28 | ||
5f09af76 DM |
29 | ifdef::wiki[] |
30 | :pve-toplevel: | |
cb84ed18 | 31 | :title: Storage |
5f09af76 DM |
32 | endif::wiki[] |
33 | ||
aa039b0f DM |
34 | The {pve} storage model is very flexible. Virtual machine images |
35 | can either be stored on one or several local storages, or on shared | |
36 | storage like NFS or iSCSI (NAS, SAN). There are no limits, and you may | |
37 | configure as many storage pools as you like. You can use all | |
38 | storage technologies available for Debian Linux. | |
39 | ||
40 | One major benefit of storing VMs on shared storage is the ability to | |
41 | live-migrate running machines without any downtime, as all nodes in | |
42 | the cluster have direct access to VM disk images. There is no need to | |
43 | copy VM image data, so live migration is very fast in that case. | |
44 | ||
8c1189b6 | 45 | The storage library (package `libpve-storage-perl`) uses a flexible |
aa039b0f DM |
46 | plugin system to provide a common interface to all storage types. This |
47 | can be easily adopted to include further storage types in future. | |
48 | ||
49 | ||
50 | Storage Types | |
51 | ------------- | |
52 | ||
53 | There are basically two different classes of storage types: | |
54 | ||
55 | Block level storage:: | |
56 | ||
57 | Allows to store large 'raw' images. It is usually not possible to store | |
58 | other files (ISO, backups, ..) on such storage types. Most modern | |
59 | block level storage implementations support snapshots and clones. | |
60 | RADOS, Sheepdog and DRBD are distributed systems, replicating storage | |
61 | data to different nodes. | |
62 | ||
63 | File level storage:: | |
64 | ||
65 | They allow access to a full featured (POSIX) file system. They are | |
66 | more flexible, and allows you to store any content type. ZFS is | |
67 | probably the most advanced system, and it has full support for | |
68 | snapshots and clones. | |
69 | ||
70 | ||
71 | .Available storage types | |
72 | [width="100%",cols="<d,1*m,4*d",options="header"] | |
73 | |=========================================================== | |
74 | |Description |PVE type |Level |Shared|Snapshots|Stable | |
75 | |ZFS (local) |zfspool |file |no |yes |yes | |
76 | |Directory |dir |file |no |no |yes | |
77 | |NFS |nfs |file |yes |no |yes | |
78 | |GlusterFS |glusterfs |file |yes |no |yes | |
79 | |LVM |lvm |block |no |no |yes | |
9801e1c3 | 80 | |LVM-thin |lvmthin |block |no |yes |yes |
aa039b0f DM |
81 | |iSCSI/kernel |iscsi |block |yes |no |yes |
82 | |iSCSI/libiscsi |iscsidirect |block |yes |no |yes | |
83 | |Ceph/RBD |rbd |block |yes |yes |yes | |
84 | |Sheepdog |sheepdog |block |yes |yes |beta | |
85 | |DRBD9 |drbd |block |yes |yes |beta | |
86 | |ZFS over iSCSI |zfs |block |yes |yes |yes | |
87 | |========================================================= | |
88 | ||
89 | TIP: It is possible to use LVM on top of an iSCSI storage. That way | |
8c1189b6 | 90 | you get a `shared` LVM storage. |
aa039b0f | 91 | |
5eba0743 FG |
92 | |
93 | Thin Provisioning | |
2afe468c | 94 | ~~~~~~~~~~~~~~~~~ |
ebc15cbc | 95 | |
8c1189b6 FG |
96 | A number of storages, and the Qemu image format `qcow2`, support 'thin |
97 | provisioning'. With thin provisioning activated, only the blocks that | |
2afe468c | 98 | the guest system actually use will be written to the storage. |
ebc15cbc | 99 | |
2afe468c | 100 | Say for instance you create a VM with a 32GB hard disk, and after |
5eba0743 | 101 | installing the guest system OS, the root file system of the VM contains |
2afe468c DM |
102 | 3 GB of data. In that case only 3GB are written to the storage, even |
103 | if the guest VM sees a 32GB hard drive. In this way thin provisioning | |
104 | allows you to create disk images which are larger than the currently | |
105 | available storage blocks. You can create large disk images for your | |
106 | VMs, and when the need arises, add more disks to your storage without | |
5eba0743 | 107 | resizing the VMs' file systems. |
2afe468c | 108 | |
8c1189b6 | 109 | All storage types which have the ``Snapshots'' feature also support thin |
2afe468c | 110 | provisioning. |
ebc15cbc | 111 | |
ba1d96fd DM |
112 | CAUTION: If a storage runs full, all guests using volumes on that |
113 | storage receives IO error. This can cause file system inconsistencies | |
114 | and may corrupt your data. So it is advisable to avoid | |
115 | over-provisioning of your storage resources, or carefully observe | |
116 | free space to avoid such conditions. | |
ebc15cbc | 117 | |
5eba0743 | 118 | |
aa039b0f DM |
119 | Storage Configuration |
120 | --------------------- | |
121 | ||
122 | All {pve} related storage configuration is stored within a single text | |
8c1189b6 | 123 | file at `/etc/pve/storage.cfg`. As this file is within `/etc/pve/`, it |
aa039b0f DM |
124 | gets automatically distributed to all cluster nodes. So all nodes |
125 | share the same storage configuration. | |
126 | ||
127 | Sharing storage configuration make perfect sense for shared storage, | |
8c1189b6 | 128 | because the same ``shared'' storage is accessible from all nodes. But is |
aa039b0f DM |
129 | also useful for local storage types. In this case such local storage |
130 | is available on all nodes, but it is physically different and can have | |
131 | totally different content. | |
132 | ||
5eba0743 | 133 | |
aa039b0f DM |
134 | Storage Pools |
135 | ~~~~~~~~~~~~~ | |
136 | ||
5eba0743 FG |
137 | Each storage pool has a `<type>`, and is uniquely identified by its |
138 | `<STORAGE_ID>`. A pool configuration looks like this: | |
aa039b0f DM |
139 | |
140 | ---- | |
141 | <type>: <STORAGE_ID> | |
142 | <property> <value> | |
143 | <property> <value> | |
144 | ... | |
145 | ---- | |
146 | ||
aa039b0f | 147 | The `<type>: <STORAGE_ID>` line starts the pool definition, which is then |
871e1fd6 FG |
148 | followed by a list of properties. Most properties have values, but some of |
149 | them come with reasonable default. In that case you can omit the value. | |
aa039b0f | 150 | |
9c41b54d DM |
151 | To be more specific, take a look at the default storage configuration |
152 | after installation. It contains one special local storage pool named | |
8c1189b6 | 153 | `local`, which refers to the directory `/var/lib/vz` and is always |
9c41b54d DM |
154 | available. The {pve} installer creates additional storage entries |
155 | depending on the storage type chosen at installation time. | |
156 | ||
8c1189b6 | 157 | .Default storage configuration (`/etc/pve/storage.cfg`) |
9801e1c3 DM |
158 | ---- |
159 | dir: local | |
aa039b0f | 160 | path /var/lib/vz |
9801e1c3 DM |
161 | content iso,vztmpl,backup |
162 | ||
9c41b54d | 163 | # default image store on LVM based installation |
9801e1c3 DM |
164 | lvmthin: local-lvm |
165 | thinpool data | |
166 | vgname pve | |
167 | content rootdir,images | |
9c41b54d DM |
168 | |
169 | # default image store on ZFS based installation | |
170 | zfspool: local-zfs | |
171 | pool rpool/data | |
172 | sparse | |
173 | content images,rootdir | |
9801e1c3 | 174 | ---- |
aa039b0f | 175 | |
5eba0743 | 176 | |
aa039b0f DM |
177 | Common Storage Properties |
178 | ~~~~~~~~~~~~~~~~~~~~~~~~~ | |
179 | ||
871e1fd6 | 180 | A few storage properties are common among different storage types. |
aa039b0f DM |
181 | |
182 | nodes:: | |
183 | ||
184 | List of cluster node names where this storage is | |
185 | usable/accessible. One can use this property to restrict storage | |
186 | access to a limited set of nodes. | |
187 | ||
188 | content:: | |
189 | ||
190 | A storage can support several content types, for example virtual disk | |
191 | images, cdrom iso images, container templates or container root | |
871e1fd6 | 192 | directories. Not all storage types support all content types. One can set |
aa039b0f DM |
193 | this property to select for what this storage is used for. |
194 | ||
195 | images::: | |
196 | ||
197 | KVM-Qemu VM images. | |
198 | ||
199 | rootdir::: | |
200 | ||
871e1fd6 | 201 | Allow to store container data. |
aa039b0f DM |
202 | |
203 | vztmpl::: | |
204 | ||
205 | Container templates. | |
206 | ||
207 | backup::: | |
208 | ||
8c1189b6 | 209 | Backup files (`vzdump`). |
aa039b0f DM |
210 | |
211 | iso::: | |
212 | ||
213 | ISO images | |
214 | ||
215 | shared:: | |
216 | ||
217 | Mark storage as shared. | |
218 | ||
219 | disable:: | |
220 | ||
221 | You can use this flag to disable the storage completely. | |
222 | ||
223 | maxfiles:: | |
224 | ||
5eba0743 | 225 | Maximum number of backup files per VM. Use `0` for unlimited. |
aa039b0f DM |
226 | |
227 | format:: | |
228 | ||
229 | Default image format (`raw|qcow2|vmdk`) | |
230 | ||
231 | ||
232 | WARNING: It is not advisable to use the same storage pool on different | |
871e1fd6 | 233 | {pve} clusters. Some storage operation need exclusive access to the |
aa039b0f | 234 | storage, so proper locking is required. While this is implemented |
871e1fd6 | 235 | within a cluster, it does not work between different clusters. |
aa039b0f DM |
236 | |
237 | ||
238 | Volumes | |
239 | ------- | |
240 | ||
241 | We use a special notation to address storage data. When you allocate | |
871e1fd6 | 242 | data from a storage pool, it returns such a volume identifier. A volume |
aa039b0f DM |
243 | is identified by the `<STORAGE_ID>`, followed by a storage type |
244 | dependent volume name, separated by colon. A valid `<VOLUME_ID>` looks | |
245 | like: | |
246 | ||
247 | local:230/example-image.raw | |
248 | ||
249 | local:iso/debian-501-amd64-netinst.iso | |
250 | ||
251 | local:vztmpl/debian-5.0-joomla_1.5.9-1_i386.tar.gz | |
252 | ||
253 | iscsi-storage:0.0.2.scsi-14f504e46494c4500494b5042546d2d646744372d31616d61 | |
254 | ||
5eba0743 | 255 | To get the file system path for a `<VOLUME_ID>` use: |
aa039b0f DM |
256 | |
257 | pvesm path <VOLUME_ID> | |
258 | ||
5eba0743 | 259 | |
aa039b0f DM |
260 | Volume Ownership |
261 | ~~~~~~~~~~~~~~~~ | |
262 | ||
8c1189b6 | 263 | There exists an ownership relation for `image` type volumes. Each such |
aa039b0f DM |
264 | volume is owned by a VM or Container. For example volume |
265 | `local:230/example-image.raw` is owned by VM 230. Most storage | |
266 | backends encodes this ownership information into the volume name. | |
267 | ||
871e1fd6 | 268 | When you remove a VM or Container, the system also removes all |
aa039b0f DM |
269 | associated volumes which are owned by that VM or Container. |
270 | ||
271 | ||
272 | Using the Command Line Interface | |
273 | -------------------------------- | |
274 | ||
871e1fd6 FG |
275 | It is recommended to familiarize yourself with the concept behind storage |
276 | pools and volume identifiers, but in real life, you are not forced to do any | |
aa039b0f DM |
277 | of those low level operations on the command line. Normally, |
278 | allocation and removal of volumes is done by the VM and Container | |
279 | management tools. | |
280 | ||
8c1189b6 FG |
281 | Nevertheless, there is a command line tool called `pvesm` (``{pve} |
282 | Storage Manager''), which is able to perform common storage management | |
aa039b0f DM |
283 | tasks. |
284 | ||
285 | ||
286 | Examples | |
287 | ~~~~~~~~ | |
288 | ||
289 | Add storage pools | |
290 | ||
291 | pvesm add <TYPE> <STORAGE_ID> <OPTIONS> | |
292 | pvesm add dir <STORAGE_ID> --path <PATH> | |
293 | pvesm add nfs <STORAGE_ID> --path <PATH> --server <SERVER> --export <EXPORT> | |
294 | pvesm add lvm <STORAGE_ID> --vgname <VGNAME> | |
295 | pvesm add iscsi <STORAGE_ID> --portal <HOST[:PORT]> --target <TARGET> | |
296 | ||
297 | Disable storage pools | |
298 | ||
299 | pvesm set <STORAGE_ID> --disable 1 | |
300 | ||
301 | Enable storage pools | |
302 | ||
303 | pvesm set <STORAGE_ID> --disable 0 | |
304 | ||
305 | Change/set storage options | |
306 | ||
307 | pvesm set <STORAGE_ID> <OPTIONS> | |
308 | pvesm set <STORAGE_ID> --shared 1 | |
309 | pvesm set local --format qcow2 | |
310 | pvesm set <STORAGE_ID> --content iso | |
311 | ||
312 | Remove storage pools. This does not delete any data, and does not | |
313 | disconnect or unmount anything. It just removes the storage | |
314 | configuration. | |
315 | ||
316 | pvesm remove <STORAGE_ID> | |
317 | ||
318 | Allocate volumes | |
319 | ||
320 | pvesm alloc <STORAGE_ID> <VMID> <name> <size> [--format <raw|qcow2>] | |
321 | ||
322 | Allocate a 4G volume in local storage. The name is auto-generated if | |
323 | you pass an empty string as `<name>` | |
324 | ||
325 | pvesm alloc local <VMID> '' 4G | |
326 | ||
5eba0743 | 327 | Free volumes |
aa039b0f DM |
328 | |
329 | pvesm free <VOLUME_ID> | |
330 | ||
331 | WARNING: This really destroys all volume data. | |
332 | ||
333 | List storage status | |
334 | ||
335 | pvesm status | |
336 | ||
337 | List storage contents | |
338 | ||
339 | pvesm list <STORAGE_ID> [--vmid <VMID>] | |
340 | ||
341 | List volumes allocated by VMID | |
342 | ||
343 | pvesm list <STORAGE_ID> --vmid <VMID> | |
344 | ||
345 | List iso images | |
346 | ||
347 | pvesm list <STORAGE_ID> --iso | |
348 | ||
349 | List container templates | |
350 | ||
351 | pvesm list <STORAGE_ID> --vztmpl | |
352 | ||
5eba0743 | 353 | Show file system path for a volume |
aa039b0f DM |
354 | |
355 | pvesm path <VOLUME_ID> | |
356 | ||
deb4673f DM |
357 | ifdef::wiki[] |
358 | ||
359 | See Also | |
360 | -------- | |
361 | ||
f532afb7 | 362 | * link:/wiki/Storage:_Directory[Storage: Directory] |
deb4673f | 363 | |
f532afb7 | 364 | * link:/wiki/Storage:_GlusterFS[Storage: GlusterFS] |
deb4673f | 365 | |
f532afb7 | 366 | * link:/wiki/Storage:_User_Mode_iSCSI[Storage: User Mode iSCSI] |
deb4673f | 367 | |
f532afb7 | 368 | * link:/wiki/Storage:_iSCSI[Storage: iSCSI] |
deb4673f | 369 | |
f532afb7 | 370 | * link:/wiki/Storage:_LVM[Storage: LVM] |
deb4673f | 371 | |
f532afb7 | 372 | * link:/wiki/Storage:_LVM_Thin[Storage: LVM Thin] |
deb4673f | 373 | |
f532afb7 | 374 | * link:/wiki/Storage:_NFS[Storage: NFS] |
deb4673f | 375 | |
f532afb7 | 376 | * link:/wiki/Storage:_RBD[Storage: RBD] |
deb4673f | 377 | |
f532afb7 | 378 | * link:/wiki/Storage:_ZFS[Storage: ZFS] |
deb4673f | 379 | |
032e755c | 380 | * link:/wiki/Storage:_ZFS_over_iSCSI[Storage: ZFS over iSCSI] |
deb4673f DM |
381 | |
382 | endif::wiki[] | |
383 | ||
251666be DM |
384 | ifndef::wiki[] |
385 | ||
aa039b0f DM |
386 | // backend documentation |
387 | ||
388 | include::pve-storage-dir.adoc[] | |
389 | ||
390 | include::pve-storage-nfs.adoc[] | |
391 | ||
392 | include::pve-storage-glusterfs.adoc[] | |
393 | ||
394 | include::pve-storage-zfspool.adoc[] | |
395 | ||
396 | include::pve-storage-lvm.adoc[] | |
397 | ||
9801e1c3 DM |
398 | include::pve-storage-lvmthin.adoc[] |
399 | ||
aa039b0f DM |
400 | include::pve-storage-iscsi.adoc[] |
401 | ||
402 | include::pve-storage-iscsidirect.adoc[] | |
403 | ||
404 | include::pve-storage-rbd.adoc[] | |
405 | ||
406 | ||
251666be | 407 | |
aa039b0f DM |
408 | ifdef::manvolnum[] |
409 | include::pve-copyright.adoc[] | |
410 | endif::manvolnum[] | |
411 | ||
251666be DM |
412 | endif::wiki[] |
413 |