mirror of
https://github.com/scylladb/scylladb.git
synced 2026-05-02 14:15:46 +00:00
Currently the loop which writes the data from the fragmented origin to the destination, moves to the next chunk eagerly after writing the value of the current chunk, if the current chunk is exhausted. This presents a problem when we are writing the last piece of data from the last chunk, as the chunk will be exhausted and we eagerly attempt to move to the next chunk, which doesn't exist and dereferencing it will fail. The solution is to not be eager about moving to the next chunk and only attempt it if we actually have more data to write and hence expect more chunks.
7.4 KiB
7.4 KiB