Details
-
Bug
-
Resolution: Fixed
-
Major
-
7.1.0
-
Untriaged
-
1
-
Unknown
Description
What's the issue?
It looks like we're currently syncing the first ~6MiB (for each vBucket) to disk multiple times, this is due to how a zero value byte range is treated by 'sync_file_range' (i.e. zero-zero is treated as the whole file).
What's the issue?
We should correctly detect/handle a zero value byte range (one which we haven't written to disk).