You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

123 lines
3.8KB

  1. [[storage_cephfs]]
  2. Ceph Filesystem (CephFS)
  3. ------------------------
  4. ifdef::wiki[]
  5. :pve-toplevel:
  6. :title: Storage: CephFS
  7. endif::wiki[]
  8. Storage pool type: `cephfs`
  9. CephFS implements a POSIX-compliant filesystem using a http://ceph.com[Ceph]
  10. storage cluster to store its data. As CephFS builds on Ceph it shares most of
  11. its properties, this includes redundancy, scalability, self healing and high
  12. availability.
  13. TIP: {pve} can xref:chapter_pveceph[manage ceph setups], which makes
  14. configuring a CephFS storage easier. As recent hardware has plenty of CPU power
  15. and RAM, running storage services and VMs on same node is possible without a
  16. big performance impact.
  17. To use the CephFS storage plugin you need update the debian stock Ceph client.
  18. Add our Ceph repository xref:sysadmin_package_repositories_ceph[Ceph repository].
  19. Once added, run an `apt update` and `apt dist-upgrade` cycle to get the newest
  20. packages.
  21. You need to make sure that there is no other Ceph repository configured,
  22. otherwise the installation will fail or there will be mixed package
  23. versions on the node, leading to unexpected behavior.
  24. [[storage_cephfs_config]]
  25. Configuration
  26. ~~~~~~~~~~~~~
  27. This backend supports the common storage properties `nodes`,
  28. `disable`, `content`, and the following `cephfs` specific properties:
  29. monhost::
  30. List of monitor daemon addresses. Optional, only needed if Ceph is not running
  31. on the PVE cluster.
  32. path::
  33. The local mount point. Optional, defaults to `/mnt/pve/<STORAGE_ID>/`.
  34. username::
  35. Ceph user id. Optional, only needed if Ceph is not running on the PVE cluster
  36. where it defaults to `admin`.
  37. subdir::
  38. CephFS subdirectory to mount. Optional, defaults to `/`.
  39. fuse::
  40. Access CephFS through FUSE, instead of the kernel client. Optional, defaults
  41. to `0`.
  42. .Configuration Example for a external Ceph cluster (`/etc/pve/storage.cfg`)
  43. ----
  44. cephfs: cephfs-external
  45. monhost 10.1.1.20 10.1.1.21 10.1.1.22
  46. path /mnt/pve/cephfs-external
  47. content backup
  48. username admin
  49. ----
  50. NOTE: Don't forget to setup the client secret key file if cephx was not turned
  51. off.
  52. Authentication
  53. ~~~~~~~~~~~~~~
  54. If you use the, by-default enabled, `cephx` authentication, you need to copy
  55. the secret from your external Ceph cluster to a Proxmox VE host.
  56. Create the directory `/etc/pve/priv/ceph` with
  57. mkdir /etc/pve/priv/ceph
  58. Then copy the secret
  59. scp cephfs.secret <proxmox>:/etc/pve/priv/ceph/<STORAGE_ID>.secret
  60. The secret must be named to match your `<STORAGE_ID>`. Copying the
  61. secret generally requires root privileges. The file must only contain the
  62. secret key itself, opposed to the `rbd` backend which also contains a
  63. `[client.userid]` section.
  64. A secret can be received from the ceph cluster (as ceph admin) by issuing the
  65. following command. Replace the `userid` with the actual client ID configured to
  66. access the cluster. For further ceph user management see the Ceph docs
  67. footnote:[Ceph user management http://docs.ceph.com/docs/luminous/rados/operations/user-management/].
  68. ceph auth get-key client.userid > cephfs.secret
  69. If Ceph is installed locally on the PVE cluster, i.e., setup with `pveceph`,
  70. this is done automatically.
  71. Storage Features
  72. ~~~~~~~~~~~~~~~~
  73. The `cephfs` backend is a POSIX-compliant filesystem on top of a Ceph cluster.
  74. .Storage features for backend `cephfs`
  75. [width="100%",cols="m,m,3*d",options="header"]
  76. |==============================================================================
  77. |Content types |Image formats |Shared |Snapshots |Clones
  78. |vztmpl iso backup snippets |none |yes |yes^[1]^ |no
  79. |==============================================================================
  80. ^[1]^ Snapshots, while no known bugs, cannot be guaranteed to be stable yet, as
  81. they lack testing.
  82. ifdef::wiki[]
  83. See Also
  84. ~~~~~~~~
  85. * link:/wiki/Storage[Storage]
  86. endif::wiki[]