This patchset removes a bug in recovery process. Current recovery process can lose data of erasure coded VDIs when a number of nodes is smaller than a number of data stripes.
The same thing can be found here: https://github.com/sheepdog/sheepdog/tree/ec-recovery Valerio, could you test it? Hitoshi Mitake (4): sheep: let add_vdi_state() be a thread safe sheep: compare oid directly sheep: rollback vnode info which is younger than last gathered epoch for ec tests/functional: a new test for erasure coding recovery sheep/recovery.c | 18 ++++++++++-------- sheep/vdi.c | 4 ++++ tests/functional/098 | 46 ++++++++++++++++++++++++++++++++++++++++++++++ tests/functional/098.out | 36 ++++++++++++++++++++++++++++++++++++ tests/functional/group | 1 + 5 files changed, 97 insertions(+), 8 deletions(-) create mode 100755 tests/functional/098 create mode 100644 tests/functional/098.out -- 1.8.3.2 -- sheepdog mailing list [email protected] http://lists.wpkg.org/mailman/listinfo/sheepdog
