Bad BRU Server Stage Mounts
There are conditions that can occur under Apple's OS X environment that can cause your BRU Server Stage archives to be written to your server system's root volume instead of the assigned Stage path on a mounted RAID or Firewire drive. The danger here is that it is possible for BRU Server to fill your root volume unexpectedly.
This condition occurs when your system has suffered from an improper shutdown or an improper disconnect of the mounted device that hosts the filesystem to which the Stage path is assigned.
This problem is related to the way OS X handles mounted volumes in the Finder. When the Finder mounts a volume the process looks like this:
Mac OS X Mounting Process:
This procedure works fine as long as the mounting and unmounting are integral and properly performed. However, when you suffer a power failure or the device is disconnected without ejecting it first, the mounted volume becomes unmounted but its mountpoint is not deleted. This means that the next mount of that device will result in the POSIX path getting the number attached. Unfortunately, BRU Server is not aware of the improper mounting and continues to write to the originally assigned Stage path. This path is now actually a directory on the server's root drive instead of the mounted volume as expected.
If/When you do lose power and suspect that this could be happening to you, bring the system back to life in single user mode (press and hold Command-s after the boot chime), manually unmount any volumes beside root that may have mounted, delete any paths in the /Volumes directory that are not assigned to the root -"/" - volume, then restart the system in normal mode. This will return mounted volumes to their correct mountpoints.
If the system does restart without your being in control of the process and BRU Server performs Staged backups, stop the BRU Server server process, copy any Stage archive files that may have been written to the root volume from that original directory (which is now on your root hard drive) to the new mountpoint Stage directory (with the number added). Unmount the volume, delete the old mountpoint and its contents (rm -rf) and then remount the volume which will reset the name of the directory without the number addition. Performing these steps will ensure that any Staged backups that were written to your root disk are available once you reset the mounting order.