Abstract: A metadirectory namespace is provided for storing files and directories “hidden” from one or more user-accessible namespaces. The metadirectory namespace is a separate, parallel namespace that coexists with other user-accessible namespaces in a storage operating system. Files and directories may be identified as being stored in a metadirectory based on their location or based on their associated flags. The metadirectory namespace may be selectively exported only to clients, users and applications given special file access permissions.

Abstract: File system is capable of performing long operations, such as deletions and truncations of long files, in a deterministic and atomic manner with regard to its recorded consistency points. The system includes both live filespace accessible to users, and user-inaccessible zombie filespace. In representative embodiments, the system includes a file deletion manager and a truncation manager capable of operating on files in the zombie filespace. Changes to the zombie filespace can be recorded in persistent memory. After a power failure or a similar event, the system can use the recorded changes to replay operations on the files in the zombie filespace, facilitating recovery. Zombie space can be sized dynamically, enabling the deletion manager to request enlargement of the zombie filespace. In some embodiments, files are transferred to the zombie filespace before links to the files' blocks in the live filespace are broken, and before performing operations on the files.

Abstract: A system and method for reallocating block in a checkpointing bitmapped-based file system is provided. The system and method enable blocks to be allocated or reallocated during checkpoints which permits various file system processes to proceed during the checkpointing process. An additional copy of the bitmap is created when the bitmap is being written to disk, thereby enabling these various file system processes to continue to operate on a bitmap during the checkpointing process.

Abstract: The present invention provides a system and method for restoring a single data stream file from a snapshot without the need to copy every individual block or inode from the snapshot. A file restore process duplicates the inode of a file within the active file system and performs a reconciliation process between the blocks of the twin inode and the snapshot inode. If the file does not exist within the active file system, a new buffer tree is created that points to the data blocks stored in the snapshot.

Abstract: The present invention provides a technique for efficiently generating and validating external file handles at a storage operating system of a filer. According to one aspect of the technique, the storage operating system associates a unique internal file handle with each file or directory stored in the filer. This unique internal file handle is used to generate a corresponding external file handle. Unlike previous approaches, the internal file handles' representations do not change depending on the location of their corresponding files or directories, e.g., on-disk or in-core, within the filer, thereby simplifying file management within the filer. According to another aspect of the inventive technique, the filer executes a file-handle validation procedure to determine whether an external file handle received with a file access request has become invalid (or “stale”) as a result of an NVRAM failure in the filer.

Abstract: The present invention provides a system and method for restoring a vdisk from a snapshot without the need to copy every individual block or inode from the snapshot. A vdisk restore process duplicates the inode of a vdisk within the active file system and performs a reconciliation process between the blocks of the twin inode and the snapshot inode. If the vdisk does not exist within the active file system, a new buffer tree is created that points to the data blocks stored in the snapshot.

Abstract: A file folding technique reduces the number of duplicate data blocks of a file consumed on a storage device of a file server configured to generate a persistent image of its active file system at a point in time. According to the file folding technique, if “old” data blocks of the file are over-written with “new” data and that new data is identical to the data of the persistent file image, the old data blocks are “freed” and the file is directed to refer to the persistent image for its data. The invention reduces disk space consumption in a file server configured to generate persistent images of its active file system and, in particular, reduces overall persistent image disk usage.

Abstract: The present invention provides a system and method for restoring a single file from a snapshot without the need to copy every individual block or inode from the snapshot. A file restore process duplicates the inode of a file within the active file system and performs a reconciliation process between the blocks of the twin inode and the snapshot inode. If the file does not exist within the active file system, a new buffer tree is created that points to the data blocks stored in the snapshot.

Abstract: In a system and method for updating a remote replicated destination file system snapshot with changes in a source file system snapshot, the snapshot being associated with a qtree (or other sub-organization of the file system volume), a qtree metadata file on the destination in provided. The qtree metadata file resides, in the destination volume, and outside the qtree structure. The metadata file stores state information with respect to each qtree. This metadata file resides at a known location in the destination volume. The metadata file can include a plurality of fixed length records having respective fields for storing various information including qtree-specific information, and can be part of a hidden metadata directory.

Abstract: In a system and method for updating a remote replicated destination file system snapshot with changes in a source file system snapshot, users and processes are redirected to a local exported snapshot of the replicated snapshot on the active file system on the destination before beginning the next update of the active file system's replicated snapshot. In this manner, an unstable replicated snapshot is not accessed. Indirection is introduced into inode lookup at the destination as the destination's active file system is being updated. The indirection can be based upon a snapshot ID that conforms to a latest exported snapshot ID.

Abstract: A system and method for mounting and unmounting volumes attached to a data storage system, such as a file server, tracks specific sub-states within each of the overall mounting and unmounting procedure states, in which specific file system requests are and are not allowed (as appropriate) with respect to the mounting/unmounting volume based upon the sub-state. More specifically, for mounting or unmounting, a storage operating system in the file server transitions a volume through a series of sub-states, as tracked in appropriate control structures. Each sub-state is characterized by a set of permitted operations that may be performed while the volume is in that sub-state, entrance and exit criteria for that sub-state, and restricted operations that may not be performed while the volume is that sub-state. During transaction request handling, the storage operating system validates each request against the sub-state to determine the disposition of the request.

Abstract: A file folding technique reduces the number of duplicate data blocks of a file consumed on a storage device of a file server configured to generate a persistent image of its active file system at a point in time. According to the file folding technique, if “old” data blocks of the file are over-written with “new” data and that new data is identical to the data of the persistent file image, the old data blocks are “freed” and the file is directed to refer to the persistent image for its data. The invention reduces disk space consumption in a file server configured to generate persistent images of its active file system and, in particular, reduces overall persistent image disk usage.

Abstract: In a system and method for updating a remote replicated destination file system snapshot with changes in a source file system snapshot, users and processes are redirected to a local exported snapshot of the replicated snapshot on the active file system on the destination before beginning the next update of the active file system's replicated snapshot. In this manner, an unstable replicated snapshot is not accessed. Indirection is introduced into inode lookup at the destination as the destination's active file system is being updated. The indirection can be based upon a snapshot ID that conforms to a latest exported snapshot ID.

Abstract: The present invention provides a system and method for restoring a single file from a snapshot without the need to copy every individual block or inode from the snapshot. A file restore process duplicates the inode of a file within the active file system and performs a reconciliation process between the blocks of the twin inode and the snapshot inode. If the file does not exist within the active file system, a new buffer tree is created that points to the data blocks stored in the snapshot.

Abstract: In a system and method for updating a remote replicated destination file system snapshot with changes in a source file system snapshot, the snapshot being associated with a qtree (or other sub-organization of the file system volume), a qtree metadata file on the destination in provided. The qtree metadata file resides, in the destination volume, and outside the qtree structure. The metadata file stores state information with respect to each qtree. This metadata file resides at a known location in the destination volume. The metadata file can include a plurality of fixed length records having respective fields for storing various information including qtree-specific information, and can be part of a hidden metadata directory.

Abstract: The invention provides a method and system for reliably performing extra-long operations in a reliable state-full system (such as a file system). The system records consistency points, or otherwise assures reliability, notwithstanding the continuous performance of extra-long operations and the existence of intermediate states for those extra-long operations. Moreover, performance of extra-long operations is both deterministic and atomic with regard to consistency points (or other reliability techniques used by the system). The file system includes a separate portion of the file system reserved for files having extra-long operations in progress, including file deletion and file truncation. This separate portion of the file system is called the zombie filespace; it includes a separate name space from the regular (“live”) file system that is accessible to users, and is maintained as part of the file system when recording a consistency point.