While the DFS Salvager performs a role similar to that of the fsck program, several major differences distinguish the two programs:
· Recovery is usually sufficient to restore file system consistency at boot time. For this reason, the Salvager is typically used only to recover, not to verify or salvage, an aggregate when the system reboots. An administrator needs to use the Salvager to salvage an aggregate only if file system damage is suspected; for example, if the log cannot be replayed successfully, if a fileset cannot be mounted, or if a controller or disk failure affects the file system.
· The Salvager does not normally prompt the issuer for direction. It asks for confirmation to proceed only if it suspects that the aggregate on which it is run is not a DCE LFS aggregate or if it finds that the size of the aggregate that is recorded on disk exceeds the capacity of the partition on which the aggregate resides. It never asks the issuer for direction on how to repair the file system, in which respect it is similar to the fsck -p command. Because of this, it can be run in the background, and several Salvager processes can be run simultaneously.
Note that the -force option of the salvage command can be used to direct the Salvager to proceed with all operations without requesting confirmation. However, if the Salvager is run on an invalid aggregate, using the -force option can produce unexpected changes.
· The Salvager displays information about files to be restored based on problems it discovers when it verifies or salvages an aggregate. A complete list of files (with path names, if possible) is printed when the operation completes. This output helps the system administrator complete the recovery of the files in the repaired file system.