MarFS provides a scalable near-POSIX file system by using one or more POSIX file systems as a scalable metadata component and one or more data stores (object, file, etc) as a scalable data component.
MarFS can ignore close_data failures under certain circumstances. This only occurs if a write never fails but the sync and/or close call on object fail. Failed writes are still caught and the object will be treated as bad. With the latest MC version this situation is extremely unlikely (but has happened for one object in our tests) because write failures short-circuit the write as a whole once the MIN_PROTECTION threshold is exceeded. Previous versions of MC did not short-circuit, so write failures were not reported until close leading to corrupt/invalid data.
MarFS can ignore
close_data
failures under certain circumstances. This only occurs if a write never fails but the sync and/or close call on object fail. Failed writes are still caught and the object will be treated as bad. With the latest MC version this situation is extremely unlikely (but has happened for one object in our tests) because write failures short-circuit the write as a whole once theMIN_PROTECTION
threshold is exceeded. Previous versions of MC did not short-circuit, so write failures were not reported until close leading to corrupt/invalid data.