OPERATIONAL DEFECT DATABASE
...
...
BugZero found this defect 1673 days ago.
One of the extents is filling up faster than the others, and you want to balance them, but only certain backups/tenants need to be moved to another extent. This article applies to Backup, Backup copy, Agent backups, Agent backups with subtenant accounts.
An extent has no free space because the Data locality placement policy has placed all files for a job on the same extent. After evacuating an extent, another extent in the SOBR has no remaining free space, and the user would like to redistribute the files manually. A new extent was added to an existing SOBR, and the user wants to manually redistribute files to the new extent.
Below are some key details to keep in mind when manually moving backup files between extents of a Scale-Out Backup Repository. (Expand each for further details.)
Please keep in mind the following regarding ReFS Fast Clone and SOBR: Due to Microsoft limitations, all backup files in the backup chain must be stored on the same volume for ReFS Fast Clone to work. For more information, see Restrictions and Remarks at Microsoft Docs. When copying files from one ReFS volume to another location, the file system downloads cloned data blocks. For this reason, copied data occupy more space in the target location than it used to occupy in the source location. This can happen, for example, if you evacuate an extent that supports block cloning from a scale-out backup repository and migrate VM backup data to another extent: copied data will require more space than it originally took.
