pvesm.adoc

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