1. 07 Feb, 2014 1 commit
  2. 11 Dec, 2013 1 commit
  3. 22 Oct, 2013 1 commit
  4. 05 Aug, 2013 1 commit
  5. 22 Jul, 2013 1 commit
    • Sofia Papagiannaki's avatar
      pithos: Fix reconcile-resource-pithos command · f415fc33
      Sofia Papagiannaki authored
      * Improved performance (reduce interaction with the database)
      * Fix command when `userid` option is provided
      
      Conflicts:
      	snf-pithos-app/pithos/api/management/commands/reconcile-resources-pithos.py
      	snf-pithos-backend/pithos/backends/lib/sqlite/node.py
      f415fc33
  6. 17 Jul, 2013 1 commit
  7. 11 Jul, 2013 2 commits
  8. 05 Jul, 2013 1 commit
  9. 04 Jul, 2013 1 commit
  10. 28 Jun, 2013 1 commit
    • Sofia Papagiannaki's avatar
      pithos: improve performance of get_domain_list · 73314a4c
      Sofia Papagiannaki authored
      Before it used to query the database
      for all the objects having the specific domain
      and finally filtered out the objects to which the user had no access.
      Now gets first the allowed_paths for the user and then
      brings those having the specific domain
      
      Refs: #3510
      73314a4c
  11. 17 Jun, 2013 1 commit
  12. 01 May, 2013 2 commits
  13. 30 Apr, 2013 1 commit
  14. 29 Apr, 2013 1 commit
  15. 22 Apr, 2013 1 commit
  16. 18 Apr, 2013 1 commit
  17. 19 Mar, 2013 2 commits
    • Sofia Papagiannaki's avatar
      Eliminate mutable default arguments · 3ad2d844
      Sofia Papagiannaki authored
      3ad2d844
    • Sofia Papagiannaki's avatar
      Fix pithos purging accounting · 0b752cff
      Sofia Papagiannaki authored
      Refs: #3442
      Refs: #1733
      
      Pithos has two different version accounting
      policies:
      * free: do not include object history version size 
      to the total account usage
      * debit: include object history version size to 
      the total account usage
      This configurable by
      PITHOS_BACKEND_FREE_VERSIONING setting.
      
      Pithos delete container function purges 
      the history versions of the including objects.
      and should send to the quota holder component
      the released disk space *only*  
      in the case of debit version accounting policy.
      
      Also the purging function used to return 
      the negative of the computed released disk space 
      but to the quotaholder component was sent 
      its negative.
      So, erroneously, the released disk space was
      added to the total quota usage.
      
      Provide tests for the following distinct cases:
      * free version accounting policy (PITHOS_BACKEND_FREE_VERSIONING=True) 
      and `auto` container versioning policy
      * free version accounting policy
      (PITHOS_BACKEND_FREE_VERSIONING=True) 
      and `none` container versioning policy
      * debit version accounting policy
      (PITHOS_BACKEND_FREE_VERSIONING=False) 
      and `auto` container versioning policy
      * debit version accounting policy(PITHOS_BACKEND_FREE_VERSIONING=False) 
      and `none` container versioning policy
      0b752cff
  18. 06 Mar, 2013 1 commit
    • Georgios D. Tsoukalas's avatar
      Fix UUID bug breaking image list and VM creation · 2e4ad3b1
      Georgios D. Tsoukalas authored
      latest_uuid(uuid) returned the latest version with the given uuid.
      This is called from pithos backend method get_uuid(),
      which is called from plankton backend method get_image().
      
      For the use in get_image() the uuid alone is not enough,
      since we need the latest version that is accessible,
      i.e. not deleted or trashed (in pithos parlance, in CLUSTER_NORMAL)
      
      The problematic scenario is cause by a MOVEing (renaming)
      a plankton image to a different name.
      Internally, pithos handles a 'move src dest' as a
      'copy src dest' + 'delete src'.
      Both copy and delete create a new version of the source,
      with the same UUID.
      We need the 'copy' version but because the delete is executed
      after the copy, the 'deleted' version becomes the latest,
      thus the one latest_uuid() would return.
      
      This has been fixed by adding a mandatory cluster argument:
      latest_uid(uuid, cluster), to which get_uuid() now specifies
      CLUSTER_NORMAL.
      The previous cluster-unaware behavior can be obtained by
      specifying cluster as None.
      
      Refs #3391
      2e4ad3b1
  19. 14 Feb, 2013 1 commit
  20. 02 Jan, 2013 1 commit
  21. 09 Nov, 2012 2 commits
  22. 13 Sep, 2012 1 commit
  23. 10 Sep, 2012 1 commit
  24. 17 Jul, 2012 2 commits
  25. 28 Jun, 2012 1 commit
  26. 26 Jun, 2012 1 commit
  27. 25 Jun, 2012 1 commit
  28. 23 Feb, 2012 1 commit
  29. 17 Jan, 2012 1 commit
  30. 12 Jan, 2012 2 commits
  31. 23 Dec, 2011 1 commit
  32. 21 Dec, 2011 1 commit
  33. 19 Dec, 2011 2 commits