when adding delegation of cloud read\writes to object io we removed the HeadObject that checks for overwrites of blocks and take it into account in the usage calculations.
since cloud resource only writes unique blocks on every write, the only way overwrites can occur are retries after failures\timeouts
Expected behavior
we should handle retries correctly and not count usage again
Steps to reproduce
somehow cause object_io to retry write_block
Screenshots or Logs or other output that would be helpful
Environment info
Actual behavior
Expected behavior
Steps to reproduce
Screenshots or Logs or other output that would be helpful
(If large, please upload as attachment)