faq.rst 5.3 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118
  1. .. _faq:
  2. .. include:: global.rst.inc
  3. Frequently asked questions
  4. ==========================
  5. Which platforms are supported?
  6. Currently Linux, FreeBSD and MacOS X are supported.
  7. You can try your luck on other POSIX-like systems, like Cygwin,
  8. other BSDs, etc. but they are not officially supported.
  9. Can I backup VM disk images?
  10. Yes, the :ref:`deduplication <deduplication_def>` technique used by |project_name|
  11. makes sure only the modified parts of the file are stored.
  12. Also, we have optional simple sparse file support for extract.
  13. Can I backup from multiple servers into a single repository?
  14. Yes, but in order for the deduplication used by |project_name| to work, it
  15. needs to keep a local cache containing checksums of all file
  16. chunks already stored in the repository. This cache is stored in
  17. ``~/.cache/borg/``. If |project_name| detects that a repository has been
  18. modified since the local cache was updated it will need to rebuild
  19. the cache. This rebuild can be quite time consuming.
  20. So, yes it's possible. But it will be most efficient if a single
  21. repository is only modified from one place. Also keep in mind that
  22. |project_name| will keep an exclusive lock on the repository while creating
  23. or deleting archives, which may make *simultaneous* backups fail.
  24. Which file types, attributes, etc. are preserved?
  25. * Directories
  26. * Regular files
  27. * Hardlinks (considering all files in the same archive)
  28. * Symlinks (stored as symlink, the symlink is not followed)
  29. * Character and block device files
  30. * FIFOs ("named pipes")
  31. * Name
  32. * Contents
  33. * Time of last modification (nanosecond precision with Python >= 3.3)
  34. * User ID of owner
  35. * Group ID of owner
  36. * Unix Mode/Permissions (u/g/o permissions, suid, sgid, sticky)
  37. * Extended Attributes (xattrs)
  38. * Access Control Lists (ACL_) on Linux, OS X and FreeBSD
  39. * BSD flags on OS X and FreeBSD
  40. Which file types, attributes, etc. are *not* preserved?
  41. * UNIX domain sockets (because it does not make sense - they are meaningless
  42. without the running process that created them and the process needs to
  43. recreate them in any case). So, don't panic if your backup misses a UDS!
  44. * The precise on-disk representation of the holes in a sparse file.
  45. Archive creation has no special support for sparse files, holes are
  46. backed up up as (deduplicated and compressed) runs of zero bytes.
  47. Archive extraction has optional support to extract all-zero chunks as
  48. holes in a sparse file.
  49. How can I specify the encryption passphrase programmatically?
  50. The encryption passphrase can be specified programmatically using the
  51. `BORG_PASSPHRASE` environment variable. This is convenient when setting up
  52. automated encrypted backups. Another option is to use
  53. key file based encryption with a blank passphrase. See
  54. :ref:`encrypted_repos` for more details.
  55. When backing up to remote encrypted repos, is encryption done locally?
  56. Yes, file and directory metadata and data is locally encrypted, before
  57. leaving the local machine. We do not mean the transport layer encryption
  58. by that, but the data/metadata itself. Transport layer encryption (e.g.
  59. when ssh is used as a transport) applies additionally.
  60. When backing up to remote servers, do I have to trust the remote server?
  61. Yes and No.
  62. No, as far as data confidentiality is concerned - all your files/dirs data
  63. and metadata are stored in their encrypted form into the repository.
  64. Yes, as an attacker with access to the remote server could delete (or
  65. otherwise make unavailable) all your backups.
  66. If a backup stops mid-way, does the already-backed-up data stay there? I.e. does |project_name| resume backups?
  67. Yes, during a backup a special checkpoint archive named ``<archive-name>.checkpoint`` is saved every 5 minutes
  68. containing all the data backed-up until that point. This means that at most 5 minutes worth of data needs to be
  69. retransmitted if a backup needs to be restarted.
  70. If it crashes with a UnicodeError, what can I do?
  71. Check if your encoding is set correctly. For most POSIX-like systems, try::
  72. export LANG=en_US.UTF-8 # or similar, important is correct charset
  73. If I want to run |project_name| on a ARM CPU older than ARM v6?
  74. You need to enable the alignment trap handler to fixup misaligned accesses::
  75. echo "2" > /proc/cpu/alignment
  76. Why was Borg forked from Attic?
  77. Borg was created in May 2015 in response to the difficulty of
  78. getting new code or larger changes incorporated into Attic and
  79. establishing a bigger developer community / more open development.
  80. More details can be found in `ticket 217
  81. <https://github.com/jborg/attic/issues/217>`_ that led to the fork.
  82. Borg intends to be:
  83. * simple:
  84. * as simple as possible, but no simpler
  85. * do the right thing by default, but offer options
  86. * open:
  87. * welcome feature requests
  88. * accept pull requests of good quality and coding style
  89. * give feedback on PRs that can't be accepted "as is"
  90. * discuss openly, don't work in the dark
  91. * changing:
  92. * Borg is not compatible with Attic
  93. * do not break compatibility accidentally, without a good reason
  94. or without warning. allow compatibility breaking for other cases.
  95. * if major version number changes, it may have incompatible changes