I came across a useful tip on the Microsoft website today, entitled “Why is placing the Sysvol directory on a separate partition a good practice?” As links like this have a habit of disappearing from the Microsoft website, I’ve reproduced the content below:
“The System Volume (Sysvol) shared directory is replicated to every domain controller in a domain by means of the File Replication Service (FRS). Here are a couple of good reasons for placing Sysvol on a separate partition:
- Sysvol’s contents and its staging files might increase in size. Placing Sysvol on a separate partition contains the growth of the directory’s contents and prevents them from consuming space on the boot partition, thereby preventing problems with other components and performance degradation.
- Placing Sysvol on its own NTFS partition minimizes disk I/O, thereby reducing the chances of receiving journal wrap errors. FRS uses the NTFS journal to monitor changes in the file system. The journal contains the update sequence number (USN) of the NTFS changes that are stored on each NTFS partition. If FRS can’t keep up with the pace of disk I/O or if FRS is turned off for a period of time, the USN that’s referenced in the FRS log might no longer exist in the NTFS volume journal. To help reduce the chance of the NTFS journal wrapping before FRS has replicated content, Windows 2000 Service Pack 3 increased the size of the NTFS journal from 32Mb 512Mb by default (with a maximum configurable limit of 10Gb).”