Open
Description
We need to have a mechanism to trigger full compaction and defragmentation through scripts. The only possible way now it by clicking it through "database main page" in web monitoring.
Also defragmentation runs only a quantum that frees up to 20 chunks, leaving all other chunks fragmented. We want to have mechanism that will consistetnly defragmentate with maximum effort.
Also it would be good if this process is controlled through BSC to prevent simultaneous compaction/defragmentation on several VDisks in the same group (in order to prevent performance degradation induced by extra processes).