[sheepdog-users] Crash of a node (maybe due to cache)

[Valerio Pachera]

I stopped the cluster, changed the fd limit, rebooted the nodes and
restarted the cluster.
Then I made another snapshot form node 'sheepdog002' that is still
shown wrong from node 'sheepdog004'.

root at sheepdog002:~# collie vdi list
  Name        Id    Size    Used  Shared    Creation time   VDI id  Copies  Tag
  zimbra_backup     0  200 GB  117 GB  0.0 MB 2013-04-16 21:41
2e519     2
s backup_data     2  1.0 TB  134 GB  890 GB 2013-06-04 18:18   c8d129
   2        second
s backup_data     3  1.0 TB  334 GB  690 GB 2013-06-05 17:29   c8d12a
   2         third
  backup_data     0  1.0 TB  0.0 MB  1.0 TB 2013-06-10 12:38   c8d12b
   2
  debian7.iso     0  621 MB  624 MB  0.0 MB 2013-06-07 14:00   ce90aa
   2
s crmdelta     2   50 GB  8.7 GB  4.8 GB 2013-06-06 08:49   e149c0
2       secondo
  crmdelta     0   50 GB  332 MB   13 GB 2013-06-06 09:25   e149c1
2
  backp        0   10 GB  580 MB  3.3 GB 2013-05-22 08:31   f313b7     2

root at sheepdog004:~# collie vdi list
  Name        Id    Size    Used  Shared    Creation time   VDI id  Copies  Tag
  zimbra_backup     0  200 GB  117 GB  0.0 MB 2013-04-16 21:41
2e519     2
  backup_data     0  1.0 TB  134 GB  890 GB 2013-06-04 18:18   c8d129
   2
s backup_data     3  1.0 TB  334 GB  690 GB 2013-06-05 17:29   c8d12a
   2         third
  backup_data     0  1.0 TB  0.0 MB  1.0 TB 2013-06-10 12:38   c8d12b
   2
  debian7.iso     0  621 MB  624 MB  0.0 MB 2013-06-07 14:00   ce90aa
   2
s crmdelta     2   50 GB  8.7 GB  4.8 GB 2013-06-06 08:49   e149c0
2       secondo
  crmdelta     0   50 GB  332 MB   13 GB 2013-06-06 09:25   e149c1
2
  backp        0   10 GB  588 MB  3.3 GB 2013-05-22 08:31   f313b7     2