A partial-state file feature capability that allows you to access the beginning of an offline file before the entire file has been recalled.
A valid daemon database entry.
See Also soft-deleted database entry and hard-deleted database entry.
The e-mail address to receive output from administrative tasks. See “Configuring the Base Object” in Chapter 2.
A floating-point constant and floating-point multiplier to use to calculate the weight given to a file's age. See “File Weighting and MSP or Volume Group Selection Parameters” in Chapter 2.
A source of additional volumes for a volume group that runs out of media. An allocation group defines a logical pool of volumes, and is different from an actual operational volume group. Normally, one allocation group is configured to serve multiple volume groups. If a volume group has an associated allocation group, when the volume group runs out of empty volumes, the library server assigns one from the allocation group to it, subject to configuration restrictions. Similarly, when a volume's hfree flag is cleared in a volume group, it is returned to the allocation group, subject to configuration restrictions. The use of allocation groups is optional. Allocation groups are defined in the DMF configuration file (/etc/dmf/dmf.conf).
The media onto which migrated data blocks are stored, usually tapes.
The combination of utilities that allows DMF to maintain a specified level of free space on a filesystem through automatic file migration.
The configuration object that defines pathname and file size parameters necessary for DMF operation.
See bit-file identifier.
See bit-file identifier.
(BFID) A unique identifier, assigned to each file during the migration process, that links a migrated file to its data on alternate media.
The collection of database entries and the user file associated with a particular bit-file identifier.
The sum of the states of the components that comprise a bit-file identifier set: the file state of any user file and the state of any database entries (incomplete, complete, soft-deleted, or active).
Physical unit of I/O to and from media, usually tape. The size of a block is determined by the type of device being written. A tape block is accompanied by a header identifying the chunk number, zone number, and its position within the chunk.
The size of I/O requests when reading from a filesystem using buffered I/O. See “Configuring Filesystems” in Chapter 2.
The floating-point constant and floating-point multiplier to use to calculate the weight given to a file's age. See:
The directory in which the volume group stores chunks while merging them from sparse tapes. See “LS Objects” in Chapter 2.
The amount of disk space (in bytes) that dmatls can use when merging chunks from sparse tapes. See “LS Objects” in Chapter 2.
The floating-point constant and floating-point multiplier to use to calculate the weight given to a file's size. See:
A list that contains an entry for each file in a filesystem eligible for migration, or for a file or range of a files that are eligible to be made offline. This list is ordered from largest file weight (first to be migrated) to smallest. This list is generated and used internally by dmfsmon(8). The dmscanfs(8) command prints similar file status information to standard output.
A catalog (CAT) record in the library server database that tracks which migrated files reside on which tape volumes.
The maximum number of child processes that the media-specific process is allowed to fork. See “Setting Up a Disk MSP” in Chapter 2.
That portion of a user file that fits on the current media (tape) volume. Most small files are written as single chunks. When a migrated file cannot fit onto a single volume, the file is split into chunks.
The binary file to execute in order to initiate a media-specific process or library server. See:
An entry in the daemon database whose path field contains a key returned by its media-specific process (MSP) or volume group, indicating that the media-specific process or volume group maintains a valid copy of the user file.
The mechanism by which data is reduced as it is written to tape.
A series of parameter definitions in the DMF configuration file that controls the way in which DMF operates. By changing the parameters associated with objects, you can modify the behavior of DMF.
A string in the DMF configuration file that defines a part of a configuration object. By changing the values associated with these parameters, you can modify the behavior of DMF. The parameter serves as the name of the line. Some parameters are reserved words, some are supplied by the site.
A program that is run automatically by the system for a specific purpose.
A database maintained by the DMF daemon. This database contains such information as the bit-file identifier, the media-specific process or volume group name, and media-specific process or volume group key for each copy of a migrated file.
The configuration object that defines parameters necessary for dmfdaemon(8) operation.
See direct-access storage device.
One or more directories into which a copy of the DMF databases will be placed. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The portion of the inode that points to the file's data blocks.
The configuration object that defines parameters for the DMF backup scripts' use of tape devices other than those defined by a drive group.
See disk cache manager.
An IBM disk drive.
The maximum size of I/O requests when using O_DIRECT I/O to read from any primary filesystem or when migrating files down the hierarchy from the STORE_DIRECTORY of a disk media-specific process running in DCM mode. See “Configuring the Base Object” in Chapter 2.
The size of I/O requests when reading from this filesystem using direct I/O. See “Configuring Filesystems” in Chapter 2.
Data on secondary storage.
The feature that lets you configure the disk media-specific process to manage data on secondary storage, allowing you to further migrate the data to tape as needed.
See drive group.
The program that accepts requests to migrate data, communicates with the operating system kernel in order to maintain a file's migration state, determines the destination of migrated data, and requests the return of offline copies.
See file state.
(DG) One of the components of an library server. The drive group is responsible for the management of a group of interchangeable tape drives located in the tape library. These drives can be used by multiple volume groups and by non-DMF processes, such as backups and interactive users. However, in the latter cases, the drive group has no management involvement; the mounting service (TMF or OpenVault) is responsible for ensuring that these possibly competing uses of the tape drives do not interfere with each other. The main tasks of the drive group are to monitor tape I/O for errors, to attempt to classify them (as volume, drive, or mounting service problems), and to take preventive action.
One or more drive groups containing drives that the library server can use for mounting and unmounting volumes. See “LS Objects” in Chapter 2.
In DCM, a cache-resident copy of a migrated file that has already been copied to tape, and can therefore be released quickly in order to prevent the cache filling, without any need to first copy it to tape (analogous to a dual-state file).
The percentage of filesystem capacity that DMF maintains as a reserve of dual-state files whose online space can be freed if free space reaches or falls below FREE_SPACE_MINIMUM. See:
A file whose data resides both online and offline.
A filesystem that has the necessary inode space to support dual-state files.
The path to a directory where a snapshot of the DMF databases will be placed when do_predump.sh is run. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The name of the drive group in the configuration file that defines how to mount the tapes that the dump tasks will use. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
One or more filesystems to dump. If not specified, the tasks will dump all the DMF-managed user filesystems configured in the configuration file. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
Specifies whether or not the dmmigrate command is run before the dumps are done to ensure that all non-dual-resident files in the DCM caches are migrated to tape. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The pathnames of one or more directories into which are copied the XFS inventory files for the backed-up filesystems. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
A maximum disk space used for files to be dumped, which may be larger or smaller than the length of the file. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
Specifies whether or not the dmmigrate command is run before the dumps are done to ensure that all migratable files in the DMF-managed user filesystems are migrated, thus reducing the number of tapes needed for the dump and making it run much faster. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The length of time that the backups of the filesystem will be kept before the tapes are reused. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The path of a file that contains tape volume serial numbers (VSNs), one per line, for the dump tasks to use. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
A file in which the VSNs of tapes that are used are written. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
Passes parameters to the xfsdump program. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
Instructs the daemon to export details of its internal request queue to SPOOL_DIR/daemon_exports every two minutes, for use by dmstat (8) and other utilities. See “Configuring the DMF Daemon” in Chapter 2.
See file handle.
An inode and its associated data blocks; an empty file has an inode but no data blocks.
The DMAPI identification for a file. You can use the dmscanfs(8), dmattr(1), and dmfind(1) commands to find file handles.
The migration state of a file as indicated by the dmattr(1) command. A file can be regular (not migrated), migrating, dual-state, offline, partial-state, unmigrating, never-migrated, or have an invalid DMF state.
A site-assigned 16-bit integer associated with a specific file allowing the file to be identified and acted upon.
Specifies whether or not dmfsmon will first free dual-state and partial-state files before freeing files it must migrate. See “DCM Policies” in Chapter 2.
Specifies whether or not dmfsmon will first free dual-state and partial-state files before freeing files it must migrate. See “Automated Space Management Parameters” in Chapter 2.
The percentage of filesystem space by which dmfsmon will decrement FREE_SPACE_MINIMUM (if it cannot find enough files to migrate) so that the value is reached. The decrement is applied until a value is found that dmfsmon can achieve. See “Automated Space Management Parameters” in Chapter 2.
The minimum percentage of free filesystem space that dmfsmon maintains. dmfsmon will begin to migrate files when the available free space for the filesystem falls below this percentage value. See “Automated Space Management Parameters” in Chapter 2.
The percentage of free filesystem space that dmfsmon will try to achieve if free space reaches or falls below FREE_SPACE_MINIMUM. See “Automated Space Management Parameters” in Chapter 2.
A user file that has been migrated and whose data blocks have been released.
A file that has one or more complete offline copies and no pending or incomplete offline copies.
The directory to use on the remote system. See “Setting Up an FTP MSP” in Chapter 2.
The Internet hostname of the remote machine on which files are to be stored. See “Setting Up an FTP MSP” in Chapter 2.
The daemon-like media-specific process (MSP) that copies data blocks onto alternate media and assigns keys to identify the location of the migrated data using the file transfer protocol (FTP) to transfer to and from disks of another system on the network.
The file containing the password to use when migrating files to the remote system. This file must be owned by root and be only accessible by root. See “Setting Up an FTP MSP” in Chapter 2.
The user name to use when migrating files to the remote system. See “Setting Up an FTP MSP” in Chapter 2.
A media-specific process or volume group database entry that has been removed from the daemon database and whose media-specific process or volume group copy has been discarded. See also active database entry and soft-deleted database entry.
The base pathname used to construct directory names for DMF directories in which databases and related files reside. See “Configuring the Base Object” in Chapter 2.
The portion of a file that contains the bit-file identifier, the state field, and the data pointers.
An entry in the daemon database for a media-specific process (MSP) or volume group that has not finished copying the data, and therefore has not yet returned a key. The path field in the database entry is NULL.
A file that has begun the migration process, but for which one or more copies on alternate media have not yet been made.
The base pathname used to construct directory names for DMF directories in which the daemon and library server database journal files will be written. See “Configuring the Base Object” in Chapter 2.
The length of time to keep journals. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The maximum size (in bytes) of the database journal file before DMF closes it and starts a new file. See “Configuring the Base Object” in Chapter 2.
(LS) The daemon-like process by which data blocks are copied onto tape and that maintains the location of the migrated data. Each library server has an associated catalog (CAT) and volume (VOL) database. A library server can be configured to contain one or more drive groups. Each drive group contains one or more volume groups. A volume group is responsible for copying data blocks onto alternate media. A volume group is capable of managing a single copy of a user file.
The full pathname of the file containing the license used by DMF. See “Configuring the Base Object” in Chapter 2.
The length of time to keep log files. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
See library server
The library servers used by the DMF daemon. See “Configuring the DMF Daemon” in Chapter 2.
The largest chunk (in bytes) that will be merged using the merge disk cache. See “LS Objects” in Chapter 2.
The maximum number of managed regions that DMF will assign to a file on a per-filesystem basis. You can set MAX_MANAGED_REGIONS to any number that is less than the actual number of regions that will fit in a filesystem attribute. See “Configuring Filesystems” in Chapter 2.
The maximum number of bytes to which the DMF daemon's virtual memory size is allowed to grow. See “Configuring the DMF Daemon” in Chapter 2.
(MSP) The daemon-like process by which data blocks are copied onto alternate media and that assigns keys to identify the location of the migrated data.
See volume merging.
The highest message level number that will be written to a log file (the higher the number, the more messages written). See:
A file that has a bit-file identifier and whose offline copies are completed. Migrated files can be dual-state or offline.
A file that has a bit-file identifier but whose offline copies are in progress.
The highest level of migration service allowed. See “Configuring the DMF Daemon” in Chapter 2, “Configuring Filesystems” in Chapter 2.
The percentage of filesystem capacity that DMF maintains as a reserve of dual-state files whose online space can be freed if free space reaches or falls below FREE_SPACE_MINIMUM. See “Automated Space Management Parameters” in Chapter 2.
Determines whether direct or buffered I/O is used when reading from this filesystem (see O_DIRECT in the open(2) man page for a description of direct I/O). See “Configuring Filesystems” in Chapter 2.
The scratch filesystem used by dmmove(8) to move files between media-specific processes or volume groups. See “Configuring the DMF Daemon” in Chapter 2.
See media-specific process.
The daemon database entry for a file that contains the path or key that is used to inform a particular media-specific process or volume group where to locate the copy of the file's data.
Names the media-specific processes used by the DMF daemon. See “Configuring the DMF Daemon” in Chapter 2.
The configuration object that defines parameters necessary for that media-specific process's operation. There is one media-specific process object for each media-specific process.
The mounting service to use. Supported values are openvault and tmf. See:
The name by which the object's devices are known to the mounting service. In the case of TMF, this is the device group name that would be used with the -g option on the tmmnt command. For OpenVault, this is the OpenVault drivegroup name that would be listed by the ov_drivegroup command. See “Configuring Device Objects” in Chapter 2, “Drive Group Objects” in Chapter 2.
A file that does not have a bit-file identifier or any offline copies. See regular file.
A file whose inode contains a bit-file identifier but whose disk blocks have been removed. The file's data exists elsewhere in copies on alternate media.
In media-specific process and volume group processing, a character string that the media-specific process or volume group returns to the daemon to indicate how a file is to be retrieved.
A storage library management facility that improves how applications can manage, store, and retrieve removable media.
An unused area in a library server catalog (CAT) database entry resulting from the removal of migrated files.
An unused database entry resulting from the removal of a migrated file during a period in which the DMF daemon is not running.
A list of access mode names that control how data is written to tape. See “Configuring Device Objects” in Chapter 2.
A list of interchange mode names that control how data is written to tape. See “Configuring Device Objects” in Chapter 2.
The file containing the OpenVault keys used by DMF. See “Configuring the Base Object” in Chapter 2.
The name returned by the hostname (1) command on the machine on which the OpenVault server is running. See “Configuring the Base Object” in Chapter 2.
See configuration parameter.
A file that has more than one region. DMF allows a file to include up to four distinct file regions. See also region .
A partial-state file feature capability that allows you to keep a specific region of a file online while freeing the rest of it (for example, if you wanted to keep just the beginning of a file online). See also partial-state file.
A partial-state file feature capability that allows you to recall a specific region of a file without recalling the entire file. For more information, see the dmput(1) and dmget(1) man pages. See also partial-state file.
Enables or disables the DMF daemon's ability to produce partial-state files. See “Configuring the DMF Daemon” in Chapter 2.
The names of the configuration objects defining policies for this filesystem. See:
Rules that tell DMF how to determine media-specific process or volume group selection, automated space-management policies, and/or file weight calculations
The configuration object that specifies parameters to determine media-specific process or volume group selection, automated space management policies, and/or file weight calculations in automatic space management.
The time in seconds since January 1, 1970.
To request that a migrated file's data be moved back (unmigrated) onto the filesystem disk, either by explicitly entering the dmget(1) command or by executing another command that will open the file, such as the vi(1) command.
The approximate rate, in seconds, at which regions of a file being recalled are put online. This allows for access to part of a file before the entire file is recalled. See “Configuring the DMF Daemon” in Chapter 2.
A contiguous range of bytes that have the same residency state. The states can be DUALSTATE, OFFLINE , MIGRATING, or UNMIGRATING.
DMF considers a regular file to be one with no bit-file identifier and no offline copies.
A tape cartridge, a tape reel, an optical disc, a digital linear tape, a removable magnetic disk, or a videotape.
A DMF maintenance command to be executed. See:
The parameter that specifies whether or not DMF will use fast scanning. If set to no, the dmscanfs command will use its recursive option, which is much slower but results in pathnames being included in the output file. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
The volume groups to use for migrating a file. (It is not used for defining which volume group to use for recalls; for that, see the definitions of the LS_NAMES, MSP_NAMES , DRIVE_GROUPS, and VOLUME_GROUPS parameters.) See:
The media-specific processes to use for migrating a file. See “File Weighting and MSP or Volume Group Selection Parameters” in Chapter 2.
The volume groups to use for migrating a file. See “File Weighting and MSP or Volume Group Selection Parameters” in Chapter 2.
A site-specific library of C++ functions that DMF will consult when making decisions about its operation.
The information about all bit-file identifier sets that is collected and analyzed by dmaudit(8). The snapshot analysis is available from the report function.
A daemon database entry for which the media-specific process or volume group copy of the data is no longer valid. Data remains on the alternate media until the database entry is hard-deleted. See also active database entry and hard-deleted database entry.
The floating-point constant and floating-point multiplier to use to calculate the weight given to a file's size. See “File Weighting and MSP or Volume Group Selection Parameters” in Chapter 2.
A tape containing only a small amount of active information.
A device file in UNIX or Linux. (DMF never migrates special files.)
The base pathname used to construct the directory names for DMF directories in which DMF log files are kept. See “Configuring the Base Object” in Chapter 2.
The field in the inode that shows the current migration state of a file.
The directory used to hold files for a media-specific process. See “Setting Up a Disk MSP” in Chapter 2.
See block.
A hardware device that reads and writes data to tape.
See chunk.
See volume merging.
A process initiated by the DMF event mechanism. Configuration tasks that allow certain recurring administrative duties to be automated are defined with configuration file parameters.
The task groups containing tasks that the daemon or library server should run. See:
The percentage of active data on a tape. DMF will consider a tape to be sparse when it has less than this percentage of data that is still active. See “Configuring Daemon Maintenance Tasks” in Chapter 2.
Command options that should be added to the tmmnt command when mounting a tape. See “Configuring Device Objects” in Chapter 2.
The base pathname used to construct the directory names for DMF directories in which DMF puts temporary files such as pipes. See “Configuring the Base Object” in Chapter 2.
The required name for the object. See:
The volume groups containing volumes that can be mounted on any of the drives within this drive group. See “Drive Group Objects” in Chapter 2.
A file that the daemon will never select as a migration candidate.
See recall.
See volume group
A bit-file-identifier (BFID) set state that consists of one or more soft-deleted daemon database entries, either incomplete or complete. There is no user file.
The process of removing the bit-file identifier (BFID) from the user file inode and soft-deleting all associated database entries.
A volume (VOL) record in the library server database that contains information about a tape volumes that exists in the pool of tapes used by the library server.
(VG) One of the components of a library server. A volume group is responsible for copying data blocks onto alternate media. Each volume group contains a pool of tapes, all of the same media type, capable of managing single copies of user files. Multiple copies of the same user files require the use of multiple volume groups. See also library server.
The mechanism provided by the library server for copying active data from volumes that contain largely obsolete data to volumes that contain mostly active data. Also known as tape merging .
The resource watcher that the library server should run. See “LS Objects” in Chapter 2.
A logical grouping of chunks. Zones are separated by file marks and are the smallest block-addressable unit on the tape volume. The target size of a zone is configurable by media type.