This statement is used for saving files and job variables to a backup archive. The user can choose between full backup, incremental backup and partial backup. For migrated files it can be specified that only the catalog entries or the catalog entries and the data are to be saved.
The backup class may be used as a criterion for deciding whether or not a file is to be backed up. Furthermore, the user can specify the storage media from and to which the data is to be written and whether the data is to be erased after backup.
An HSMS administrator can assign the files or job variables a different user ID or catalog ID.
Archive attributes can be set for BACKUP-FILES requests which specify that obsolete backups are to be deleted implicitly.
This statement has different formats for nonprivileged users and HSMS administrators. Privileged operands or operand values are marked with *P).
Format
BACKUP-FILES | Alias: BCF | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*P) Privileged operand or operand value
FILE-NAMES =
Serves to specify the files to be backed up. Specification of this operand is mandatory. The user can further restrict the selection made here by means of the operands EXCEPT-FILE-NAMES, SUPPORT, SELECT-FILES and MAXIMUM-BACKUP-CLASS.
If files from more than one SF pubset are specified and are to be saved to a default system archive, then the same default system archive must be assigned to all these pubsets. Otherwise, the statement will be rejected. To prevent this you should only specify files from one pubset for each backup request.
All nonprivileged users can also archive files belonging to other user IDs if they are co-owners of these files. They can back up the files in one of their own archives or in an archive belonging to the file owner if they are also co-owners of the archive.
FILE-NAMES = *OWN
All files of the user’s own ID residing on any imported pubset (except for shared SF pubsets for which the home computer is the slave) in the specified environment are to be backed up.
FILE-NAMES = *ALL
Meaning for an HSMS administrator:
In the specified environment, all files residing on any imported pubset (except for shared SF pubsets for which the home computer is the slave) are to be backed up.
Meaning for all other users:
*ALL has the same effect as *OWN.
FILE-NAMES = *NONE
No files are to be backed up.
FILE-NAMES = *SELECTED
The path names of the files to be backed up are to be taken from a list that was compiled within the same HSMS run by means of the HSMS statement SELECT-FILE-NAMES.
The files to be processed must be available in the specified environment.
FILE-NAMES = *FROM-FILE(...)
The path names of the files to be backed up are to be taken from a file. The nonprivileged caller must be owner or co-owner of this file. This list file must be a SAM file with variable-length records containing one path name per record. Only upper-case characters may be used. The list file can be created, for instance, by means of the HSMS statement SELECT-FILE-NAMES or the DMS command SHOW-FILE-ATTRIBUTES.
The files to be processed must be available in the specified environment.
LIST-FILE-NAME = <filename 1..54 without-gen-vers>
Path name of the list file.
FILE-NAMES = *FROM-LIBRARY-ELEMENT(...)
The path names of the files which are to be backed up are taken from a PLAM library element (type S). The library element contains one path name per record. Only upper-case letters may be used.
LIBRARY = <filename 1..54 without-gen-vers>
Name of the PLAM library.
ELEMENT = <composed-name 1..64 with-under>
Name of the type-S element. The element of the highest existing version is used.
FILE-NAMES = list-poss(20): <filename 1..80 without-vers-with-wild> / <partial-filename 2..79 with-wild>
The path names of the files to be backed up are specified directly. A list of up to 20 names may be specified.
The file names may be specified as fully or partially qualified names, with or without a catalog/user ID. If required, the file name is extended by the user ID of the request and the catalog ID which is taken either from the specified environment or from the default catalog ID of the user ID.
The files can alternatively be selected using wildcards. Note, however, that only the HSMS administrator may use wildcards for user ID specification because only the HSMS administrator is allowed to specify other user IDs. In the case of shared pubsets, the replacement of wildcards for catalog IDs does not refer to SF pubsets for which the home computer is the slave.
The following applies to file generations: when specifying fully qualified file names, the specification of a generation number (version) is permissible. Only the specified generations will be backed up.
EXCEPT-FILE-NAMES =
Serves to specify files that are to be excluded from backup.
EXCEPT-FILE-NAMES = *NONE
All files specified with the FILE-NAMES operand are to be backed up.
EXCEPT-FILE-NAMES = *FROM-FILE(...)
The path names of the files to be excluded from backup are to be taken from a file. The nonprivileged caller must be owner or co-owner of this file. This list file must be a SAM file with variable-length records containing one path name per record. File names must be in uppercase characters only. The list file can be created, for instance, by means of the SELECT-FILE-NAMES statement or the DMS command SHOW-FILE-ATTRIBUTES.
LIST-FILE-NAME = <filename 1..54 without-gen-vers>
Path name of the list file.
EXCEPT-FILE-NAMES = *FROM-LIBRARY-ELEMENT(...)
The path names of the files which are not to be backed up are taken from a PLAM library element (type S). The library element contains one path name per record. Only upper-case letters may be used.
LIBRARY = <filename 1..54 without-gen-vers>
Name of the PLAM library.
ELEMENT = <composed-name 1..64 with-under>
Name of the type-S element. The element of the highest existing version is used.
EXCEPT-FILE-NAMES = list-poss(20): <filename 1..80 without-vers-with-wild> /
<partial-filename 2..79 with-wild>
The path names of the files to be excluded from backup are specified directly. A list of up to 20 names may be specified.
The first character of the file names must not be a hyphen. The file names may be specified as fully or partially qualified names, with or without a catalog/user ID. If required, the file name is extended by the user ID of the request and the catalog ID, which is either taken from the specified environment or from the default catalog ID of the user ID.
You can also use wildcard syntax to select the files.
In the case of shared pubsets, the replacement of wildcards for catalog IDs does not refer to SF pubsets for which the home computer is the slave.
The following applies to file generations: when specifying fully qualified file names, the specification of a generation number (version) is permissible. Only the specified generations will be excluded from backup.
NEW-FILE-NAMES =
This operand is only available to the HSMS administrator.
The HSMS administrator can back up files, simultaneously assigning them a new name. This facilitates subsequent relocation scenarios. The new file names have the following format:
:<cat-id>:$<user-id>.old-filename-without-cat-user
NEW-FILE-NAMES = *SAME
The files are backed up under their original file names.
NEW-FILE-NAMES = *BY-RULE(...)
The files are assigned new names according to a specific rule. This operand is valid only if SELECT-FILES=*ALL-FILES is specified.
NEW-CATALOG-ID =
Specifies whether the files are to be backed up under another catalog ID.
NEW-CATALOG-ID = *SAME
The files are backed up under their current catalog ID.
NEW-CATALOG-ID = <cat-id>
The files are backed up under the specified catalog ID. The specified catalog ID must agree with the backup environment. The catalog ID must be specified without colons. The user ID must be entered in this catalog.
NEW-USER-ID =
Specifies whether the files may be backed up under another user ID.
NEW-USER-ID = *SAME
The files are backed up under their current user ID.
NEW-USER-ID = <name 1..8>
The files are backed up under the specified user ID. The user ID must be specified without the leading dollar sign.
ENVIRONMENT =
Defines the HSMS environment in which the HSMS statement is to be processed.
ENVIRONMENT = *STD
Meaning for privileged users: *STD is set to the value *SINGLE-FEATURE.
Meaning for nonprivileged users: *STD is the environment that is associated with the user’s default pubset (the default pubset is defined in the user catalog).
ENVIRONMENT = *SINGLE-FEATURE
The HSMS statement is valid in the SF environment.
ENVIRONMENT = *SYSTEM-MANAGED(...)
The HSMS statement is valid in the specified SM pubset environment.
CATALOG-ID = <cat-id>
Catalog ID of the SM pubset for which the HSMS statement is valid.
MANAGEMENT-CLASS =
This operand is only available to the HSMS administrator.
The objects to be processed are selected in accordance with the management class assigned to them.
MANAGEMENT-CLASS = *NONE
The HSMS statement is processed with all specified objects to which no management class is assigned.
MANAGEMENT-CLASS = *ANY
The HSMS statement is processed irrespective of whether the objects are assigned to a management class or not.
MANAGEMENT-CLASS = <alphanum-name 1..8>
The HSMS statement is processed with all the specified objects to which the specified management class is assigned.
SELECT-FILES =
Serves to define the scope of the backup. In addition to a full backup, an incremental backup or a partial backup, backup can also only be performed for files which were modified after a particular date.
SELECT-FILES = *MODIFIED-FILES(...)
An incremental backup is to be carried out, i.e. files will be backed up only if their current version is not yet present in the specified archive.
PARTIAL-FILE-SAVE =
The incremental backup can be further restricted to a partial backup, which means that of the files specified here only those blocks are to be saved which have been modified since the last full backup.
Up to 255 partial backups are permitted between any two full backups.
A save file cannot be continued by a partial backup, if the save file contains more than one save version (SAVE-FILE-STRUCTURE=*SEVERAL-SVID).
PARTIAL-FILE-SAVE = *NO
No partial backup is to be performed.
PARTIAL-FILE-SAVE = *LARGE-FILES
The files specified as LARGE in the catalog are to be partially saved. All other files are fully saved.
PARTIAL-FILE-SAVE = *YES (MINIMUM-SIZE=<integer 0..2147483647 2Kbyte >)
The files which have at least the specified size are to be partially saved. All other files are fully saved.
SELECT-FILES = *ALL-FILES(...)
A full backup is to be performed, i.e. all files are to be completely saved. The subordinate operand FROM is only available to the HSMS administrator.
FROM = *S0
All files on storage level S0 are to be backed up, i.e. the entire backup is performed online.
This is the preset value for nonprivileged users.
Specification of this operand value is mandatory if the HSMS administrator is renaming files/job variables (NEW-FILE-NAMES/NEW-JV-NAMES operand).
FROM = *LATEST-BACKUPS-OR-S0
Modified files are to be backed up from storage level S0.
Files that have not been modified are to be backed up from their last full or incremental backup in the current archive.
FROM = *ONLY-LATEST-BACKUPS
All files which refer to the latest save version are to be backed up from their last full or incremental backup in the current archive.
SELECT-FILES = *BY-CATALOG-MODIFICATION(...)
An incremental backup is only performed on the basis of the catalog entries: only those files are backed up whose catalog entry was modified after a certain date.
As archive entries are not taken into account, this backup takes place independently of previous backups. Note that files that have been modified before the specified date and since then have not been backed up are not backed up.
Job variables, group entries of file generation groups and - as far as possible - open files are saved regardless of the modification date.
For save versions created with this setting, the backup scope is displayed in the output of the SHOW-ARCHIVE statement with “CAT-F”. The save version contains only “fully saved” files and job variables; catalog entries for files that are not saved (save type CNS) are not included. Consequently the recovery of large data volumes should not only be restricted to one save version, but all save versions since the last full backup should be specified.
CHANGED-AFTER =
All files are taken into account which have been modified since the specified date. With explicit date specification the modification time can be defined more precisely by specifying the time.
CHANGED-AFTER = *LATEST-SAVE-VERSION-DATE
Date and time of the last save are used, in other words all files are taken into account whose catalog entries have been modified since this save. This setting can be used if saves to this archive always incorporate the same file set and the last save was completed successfully.
CHANGED-AFTER = <date with-compl>(...)
Explicit specification of the modification date.
TIME = 00:00:00 / <time>
Time that determines the modification time specifically.
For files and private disks any explicit time specification is ignored and the default TIME=00:00:00 assumed.
CHANGED-AFTER = <integer -99999..0 days >(...)
Explicit specification of the modification date in days relative to the current date.
TIME = 00:00:00 / <time>
Time that determines the modification time specifically.
For files and private disks any explicit time specification is ignored and the default TIME=00:00:00 assumed.
SUPPORT =
The files to be saved are selected via the type of volume (public disk, private disk or tape) on which they reside.
SUPPORT = *ANY
The type of volume on which the files reside is not a selection criterion.
SUPPORT = *PUBLIC-DISK(...)
Restricts backup to files residing on the disks of a pubset or on the assigned Net-Storage.
STORAGE-TYPE =
Selects the storage type for public volumes.
STORAGE-TYPE = *ANY
The files are saved irrespectively of the storage type. They can reside on the disks of a pubset or on Net-Storage. Files residing on Net-Storage volumes of NETVOL volume type are saved only on BS2000 V21.0A or higher.
STORAGE-TYPE = *PUBLIC-SPACE
Only files which reside on the disks of a pubset are saved. Files on Net-Storage are ignored.
STORAGE-TYPE = *NET-STORAGE(...)
Only files which reside on Net-Storage volumes are saved.
VOLUMES = *ALL / list-poss(150): <vsn 1..6>
Specifies the Net-Storage volume on which the files to be saved reside. If more than one Net-Storage is assigned to the pubset, the Net-Storage volumes from which files are to be saved can be specified in a list. *ALL selects all Net-Storage volumes which are assigned to the specified pubsets.
FILE-TYPE =
Selects the file type of Net-Storage files that will be saved.
FILE-TYPE = *ANY
The Net-Storage files are selected irrespective of the file type.
FILE-TYPE = *BS2000
Only Net-Storage files of the type BS2000 are selected.
FILE-TYPE = *NODE-FILE
Only Net-Storage files of the type node file are selected.
SUPPORT = *PRIVATE-DISK(...)
Restricts backup to files which reside on private disks and which have been imported, i.e. recorded in the catalog of a pubset.
VOLUMES = *ALL
All files located on private disks are saved.
VOLUMES = list-poss(150): <vsn 1..6>
Volume serial numbers of private disks. All files having at least one extent on the specified disks are backed up.
SUPPORT = *TAPE
Only the catalog entries of tape files are backed up.
MAXIMUM-BACKUP-CLASS = *STD / *A / *B / *C / *D / *E
Only the files whose backup class is equal or inferior to the backup class specified here are to be backed up. *E saves files of all backup classes.
Meaning of *STD:
If a new save file is created, the default value set for ARCHIVE is used (see the “ARCHIVE” manual [2], SAVE statement).
If a save file is continued, it is continued with the same backup class with which it was created. Save files which can only contain one save version (SINGLE-SVID) can only be continued with the same backup class with which they were created.
SAVE-OPTIONS =
Save options can be defined during the course of this backup run.
SAVE-OPTIONS = *STD
The default values specified for each operand described below apply to the following save options.
SAVE-OPTIONS = *PARAMETERS(...)
The following save options can be modified:
SAVE-DATA =
Specifies whether just the catalog entry of migrated files or both the catalog entry and the data are to be saved.
SAVE-DATA = *STD
The archive’s standard setting applies.
SAVE-DATA = *S0
Data is to be saved only for files on S0. For migrated files, only the metadata is saved.
SAVE-DATA = *S1-S0
Data is to be saved only for files on S0 and files migrated to S1.For files migrated to S2, only the metadata is saved.
SAVE-DATA = *S2-S1-S0
Data is to be saved for all files, regardless of whether they have been migrated or not.Specification of this operand value is mandatory if the HSMS administrator is renaming files/job variables (NEW-FILE-NAMES/NEW-JV-NAMES operand).
SAVE-ACL = *YES / *NO
This operand only exists for compatibility reasons and is not evaluated.
SAVE-ONLINE-FILES = *NO / *YES
Determines whether or not appropriately marked database files are to be backed up even when open (online).
SAVE-DIRECTORY = *NO / *YES
Determines whether the archive directory used for this run is written to the backup volume as the last file in the run.
If the archive directory is included in the backup, its name remains unchanged even if files are renamed during backup (NEW-FILE-NAMES operand).
The VSN of the volume to which the archive directory was saved is listed in the report, also in a summary report (REPORT=*SUMMARY).
SAVE-PLAM-INFO = *STD / *NO / *YES
Determines whether information about the element structure should also be written to the save volume when PLAM library files are saved. Only with this additional information can elements of a library file be restored with the RESTORE-LIBRARY-ELEMENTS statement. If this information is not contained on the save volume, a library can only be restored as a complete library file.
The library files saved with the element structure are listed in the full report (REPORT=*FULL) of the save type FULB or PARB (see also SHOW-ARCHIVE output in section "SHOW-ARCHIVE Show archive directory").
Note
The element structure of migrated PLAM libraries cannot be saved.
SAVE-PLAM-INFO = *STD
The default setting for the archive applies.
SAVE-NET-STOR-DATA = *YES / *NO
Determines whether Net-Storage files residing on NETSTOR volume type of BS2000 file type are to be saved in their entirety (data and catalog entries).
When *NO is specified, only their catalog entries are saved. The save type in this case is CATL.
Note
Net-Storage files of the type node file are always saved in their entirety (data and catalog entries). Net-Storage files residing on NETVOL volume type are always saved with data and catalog entries.
SAVE-SAM-STRUCTURE = *NO/*YES
Determines whether the structure of SAM-Node files is to be saved.
SAM node files that have been saved without SAM structure can only be restored as a SAM node file again.
SAM node files that have been saved with SAM structure can be restored either as FILE-TYPE=*BS2000 (on public space or Net-Storage) or FILE-TYPE=*NODE-FILE.
JV-NAMES =
Serves to specify the job variables to be backed up.
If job variables from more than one SF pubset are specified and are to be saved to a default system archive, the same default system archive must be assigned to all these pubsets. Otherwise, the statement will be rejected. To prevent this you should only specify job variables from one pubset for each backup request.
All nonprivileged users can also archive job variables belonging to other user IDs if they are co-owners of these job variables. They can back up the job variables in one of their own archives or in an archive belonging to owner of the job variables.
JV-NAMES = *NONE
No job variables are to be backed up.
JV-NAMES = *OWN
In the specified environment, all job variables of the user’s own ID residing on any imported pubset (except for shared SF pubsets for which the home computer is the slave) are to be backed up.
JV-NAMES = *ALL
All job variables residing on all imported pubsets (except for shared SF pubsets for which the home computer is the slave) in the specified environment are to be backed up.
JV-NAMES = *SELECTED
The path names of the job variables to be backed up are taken from a list which is created in the same HSMS run with the HSMS statement SELECT-JV-NAMES.
JV-NAMES = *FROM-FILE(...)
The path names of the job variables to be backed up are to be taken from a file. The nonprivileged caller must be owner or co-owner of this file. This list file must be a SAM file with variable-length records containing one path name per record. Only upper-case characters may be used.
The job variables to be processed must be available in the specified environment.
LIST-FILE-NAME = <filename 1..54 without-gen-vers>
Path name of the list file.
JV-NAMES = *FROM-LIBRARY-ELEMENT(...)
The path names of the job variables which are to be backed up are taken from a PLAM library element (type S). The library element contains one path name per record. Only upper-case letters may be used.
LIBRARY = <filename 1..54 without-gen-vers>
Name of the PLAM library.
ELEMENT = <composed-name 1..64 with-under>
Name of the type-S element. The element of the highest existing version is used.
JV-NAMES = list-poss(20): <filename 1..80 without-vers-with-wild> / <partial-filename 2..79 with-wild>
The path names of the job variables to be backed up are specified directly. A list of up to 20 names may be specified.
The job variables to be processed must be available in the specified environment.
The job variables can alternatively be selected using wildcards. In the case of shared pubsets, the replacement of wildcards for catalog IDs does not refer to SF pubsets for which the home computer is the slave.
If job variables of a shared pubset imported in slave mode are specified, no other public volume set of a different master may be specified.
EXCEPT-JV-NAMES =
Serves to specify job variables that are to be excluded from backup.
EXCEPT-JV-NAMES = *NONE
All job variables specified in the JV-NAMES operand are to be backed up.
EXCEPT-JV-NAMES = *FROM-FILE(...)
The path names of the job variables to be excluded from backup are to be taken from a file. The nonprivileged caller must be owner or co-owner of this file. This list file must be a SAM file with variable-length records containing one path name per record. Only upper-case characters may be used.
LIST-FILE-NAME = <filename 1..54 without-gen-vers>
Path name of the list file.
EXCEPT-JV-NAMES = *FROM-LIBRARY-ELEMENT(...)
The path names of the job variables which are not to be backed up are taken from a PLAM library element (type S). The library element contains one path name per record. Only upper-case letters may be used.
LIBRARY = <filename 1..54 without-gen-vers>
Name of the PLAM library.
ELEMENT = <composed-name 1..64 with-under>
Name of the type-S element. The element of the highest existing version is used.
EXCEPT-JV-NAMES = list-poss(20): <filename 1..80 without-vers-with-wild> /
<partial-filename 2..79 with-wild>
The path names of the job variables to be excluded from backup are specified directly. A list of up to 20 names may be specified.
The first character of the job variable names must not be a hyphen.
The job variables can alternatively be selected using wildcards. In the case of shared pubsets, the replacement of wildcards for catalog IDs does not refer to SF pubsets for which the home computer is the slave.
NEW-JV-NAMES =
This operand is only available to the HSMS administrator.
The HSMS administrator can back up job variables, simultaneously assigning them a new name. This facilitates subsequent reorganization. The new job variable names have the following format:
:<cat-id>:$<user-id>.old-jvname-without-cat-user
NEW-JV-NAMES = *SAME
The job variables are backed up retaining their original names.
NEW-JV-NAMES = *BY-RULE(...)
The job variables are backed up under new names. This operand is valid only if SELECT-FILES=*ALL-FILES is specified.
NEW-CATALOG-ID =
The job variables may be backed up under another catalog ID.
NEW-CATALOG-ID = *SAME
The job variables are backed up under their current catalog ID.
NEW-CATALOG-ID = <cat-id>
The job variables are backed up under the specified catalog ID. The specified catalog ID must agree with the backup environment. The catalog ID must be specified without colons.
The user ID must have an entry in this catalog.
NEW-USER-ID =
The job variables may be backed up under another user ID.
NEW-USER-ID = *SAME
The job variables are backed up under their current user ID.
NEW-USER-ID = <name 1..8>
The job variables are backed up under the specified user ID. The user ID must be specified without the leading dollar sign.
DELETE-FILES-AND-JV = *NO / *YES(...)
Defines whether the backed up files and job variables are to be deleted from processing level S0 once they have been written to the archive.
In the case of automatic duplication to a shadow archive, the files are not deleted until the duplication is completed.
DELETE-FILES-AND-JV = *YES(...)
The backed up files and job variables are to be deleted from S0.
Exception
Files and job variables of the SYSHSMS ID are excluded from deletion.
Deletion can be restricted by specifying which file and job variable protection attributes are to be respected:
PROTECTION-RESPECTED = *ALL
Deletion is to be restricted to files and job variables without password protection, with write access and whose retention period has expired.
PROTECTION-RESPECTED = *PASSWORDS
Deletion is to be restricted to files and job variables without password protection.
PROTECTION-RESPECTED = *NONE
This operand value is only available to the HSMS administrator.
All files and job variables are to be deleted, irrespective of their protection attributes.
ARCHIVE-NAME = *SYSBACKUP / <filename 1..22 without-cat-gen-vers>
Name of the archive to which the files and job variables are to be saved. The specified archive must already exist in the specified environment and be available for use as a backup archive.
If the archive directory is protected by a password, this password must be entered by means of the ADD-PASSWORD command prior to statement entry. This also applies to HSMS administrators.
Specification of *SYSBACKUP as the archive name is restricted to the HSMS administrator.
SAVE-FILE =
Defines the save file to which the backed up files are to be written.
SAVE-FILE = *NEW(...)
The backed up files are to be written to a new save file.
Specification of this operand value is mandatory if the HSMS administrator is renaming files/job variables (NEW-FILE-NAMES/NEW-JV-NAMES operand).
The following attributes for this save file can be defined here:
RETENTION-PERIOD = *STD / <integer 0..16383 days >
(Physical) retention period in days. During this period, neither the save file nor the save volume may be modified or deleted.
If the HSMS statement is used with a management class, *STD means that the operand value is taken from the corresponding attribute of the management class.
If the HSMS statement is used without a management class, *STD means that the value is taken from the archive definition.
USER-ACCESS = *OWNER-ONLY
The save file is to be created as non-shareable. This implicitly protects the save file against access by other users who are working without HSMS.
USER-ACCESS = *ALL-USERS
The save file is to be created as shareable, in other words other users can also access it.
If the corresponding archive directory is shareable and resides under TSOS, each user is simultaneously enabled to restore his or her files using ARCHIVE.
SAVE-FILE = *CONTINUE(...)
If the archive definition permits no more than one save version per save file (see the HSMS statement CREATE-ARCHIVE, SAVE-FILE-STRUCTURE=*SINGLE-SVID), the specified save file is to be extended by a disjunct set of files. Continuation of the save file of such backup archives implies extension of the save version it contains.
If such a save file is to be continued, the operands MAXIMUM-BACKUP-CLASS and SELECT-FILES must have the same values as when the save file was created. Furthermore, the save file must be continued using the same storage level and the same device type.
If the archive definition permits more than one save version per save file (SAVE-FILE-STRUCTURE=*SEVERAL-SVID), a new save version is added to the specified save file.
Note
Even if the archive definition permits several save versions for each save file, only save files on storage level S2 can contain more than one save version.
A save file created as *SINGLE-SVID cannot be continued once the archive is changed from *SINGLE-SVID to *SEVERAL-SVID.
Save file continuation is prohibited if files are to be selected with SELECT-FILES= *ALL-FILES(FROM=*LATEST-BACKUPS-OR-S0).
If the last SFID/SVID is not consistent due to a DMS error the current save file cannot be continued. Therefore a new save file for this directory must be created.
It is important to bear in mind when saving to storage level S1, that save files can only be continued under the following conditions: SAVE-FILE-STRUCTURE of the archive is set to *SINGLE-SVID. In addition ...:
The SAVE-FILE-PROCESSING and S1-VOLUME-SET settings under which the original backup was created must be identical to the current settings under which the save file is to be continued. In the case of SAVE-FILE-PROCESSING, the same value must be set, either
*HSMS-V10-COMPATIBLE or *HSMS-V9-COMPATIBLE. A save file can only be continued if the value of the operand SAVE-SAM-STRUCTURE is the same as that of the save file to be continued.
SAVE-FILE-ID = *LATEST
The most recently created save file for this archive is to be continued.
SAVE-FILE-ID = <composed-name 15..15>
The specified save file is to be continued. The save file ID must be specified as follows: S.yymmdd.hhmmss
If this specification is made for an archive with *SEVERAL-SVID structure, only the last save file in the archive can be processed.
SAVE-FILE = *STD
A standard save file is to be created or continued. The (physical) retention period valid for the save file is defined in the archive definition.
This operand value may be specified only for archives that permit more than one save version per save file (see the HSMS statement CREATE-ARCHIVE, SAVE-FILE-STRUCTURE=*SEVERAL-SVID).
SAVE-VERSION-NAME = *NONE / <name 1..8>
Name of the save version generated by the backup request. This name can be used to refer to the save version for restore operations or in the HSMS statement SHOW-ARCHIVE. The name is extended internally by a prefix derived from the user ID (or, in the case of the HSMS administrator, by SYSHSMS). Any save version is uniquely identified by its name and an internal time stamp.
If the archive definition permits no more than one save version per save file (SAVE-FILE-STRUCTURE=*SINGLE-SVID), any save version name specified for save file continuation will be ignored. Unless otherwise specified, no save version name is assigned.
COMPRESS-FILES = *STD / *YES / *NO
Determines whether or not data is compressed before being written to the volume.
Unless otherwise specified, the default value in the archive definition is used. The operand is not evaluated when saving to magnetic tape cartridge as the data in this case might be compressed by the device.
TO-STORAGE =
Defines the storage level at which the files and job variables are to be backed up. Private disks and pubsets are permitted in addition to levels S1 and S2. Save files must be continued using the same storage level and the same device type as for their creation.
TO-STORAGE = *S2-STORAGE-LEVEL(...)
The files are to be backed up to storage level S2 of the specified environment. The volumes to be used can be specified. If more than one volume is specified, all volumes must be of the same device type.
VOLUMES = *FROM-POOL
The volumes are to be taken from the volume pool of the specified archive, i.e. the pool of the archive directory or the associated MAREN pool.
VOLUMES = *FROM-OPERATOR
The volumes are to be allocated by the operator on request.
VOLUMES = list-poss(20): <vsn 1..6>
List of volume serial numbers, which will be requested in the specified order.
DEVICE-TYPE = *STD / <device>
Device type of the requested volumes. The device type must belong to the class “TAPE”. Only device types known in the system are accepted. In interactive mode, DEVICE-TYPE=? calls up a list of the available device types.
Unless otherwise specified, the preset value from the archive definition applies if a new save file is created (S2-DEVICE-TYPE). If an existing save file is continued, the device type specified when the save file was created is used.
LOCATION = *STD / *ANY / <alphanum-name 1..8>
Location used for requesting volumes. This location must be known to MAREN. If the location manager is in use, the location should be properly defined in MARENLM. Otherwise MAREN ignores the specified value.
If MAREN is not in use, you must specify for LOCATION the value *ANY, or *STD if *ANY is the default location of the archive.
LOCATION = *STD
The default location of the archive is used.
LOCATION = *ANY
No location is to be used.
TO-STORAGE = *S1-STORAGE-LEVEL
This value applies in an SF pubset environment, as well as in an SM pubset environment. In an SF pubset environment, the files are stored on an S1 pubset. If the global HSMS parameter SAVE-FILE-PROCESSING is set to *HSMS-V10-COMPATIBLE this may also be an S1-SM-pubset. All pubsets from which files are stored must have access to the same S1 pubset.
In an SM pubset environment, the files are stored on the S1 volume set. When an extended S1 level is defined for the SM pubset (S1-VOLUME-SET=*ALL-HSMS-CONTROLLED), TO-STORAGE=*S1-STORAGE-LEVEL is only accepted under the following circumstances:
SAVE-FILE-PROCESSING=*HSMS-V10-COMPATIBLE is set.
BS2000 OSD/BC V11.0 or higher is used on the system. In the case of a shared pubset network, all pubset sharers of the SM pubset must satisfy this requirement.
When SAVE-FILE-PROCESSING=*HSMS-V9-COMPATIBLE is set, saving to S1 level is possible only if a volume set is specified explicitly.
TO-STORAGE = *PRIVATE-DISK(...)
This value is only valid in an SF pubset environment.
The files are to be backed up to private disk. The private disks to be used can be specified. If more than one private disk is specified, all of them must be of the same device type.
At least one private disk must be specified for each parallel run, i.e. the number of specified archive numbers must be greater than or equal to the number of parallel runs.
VOLUMES = list-poss(150): <vsn 1..6>
List of volume serial numbers of the disks to which the files are to be backed up.
DEVICE-TYPE = STDDISK / <device>
Device type of the private disks. The default setting STDDISK specifies that the default device type is to be used. Only device types known in the system are accepted. In interactive mode, DEVICE-TYPE=? calls up a list of the available device types.
TO-STORAGE = *PUBLIC-DISK(...)
This operand value is only available to the HSMS administrator.
This value is only valid in an SF pubset environment.
The files are backed up to a pubset.
PUBSET-ID = <cat-id>
Catalog ID of the pubset.
TO-STORAGE = *NET-STORAGE(...)
This value is only valid in an SF pubset environment.
The files are backed up to Net-Storage.
VOLUMES = list-poss(16): <vsn 1..6>
Specifies the Net-Storage volume to which the files are backed up. The specified volume must be accessible.
More than one Net-Storage volume can also be specified in a list (e.g. when there is more than one save task).
CONCURRENT-COPY =
Specifies whether this function is to be activated, i.e. whether modifications can be made to files during the backup (following initialization of the Concurrent Copy session).
CONCURRENT-COPY = *NO
The Concurrent Copy function is not to be activated.
CONCURRENT-COPY = *YES(...)
The Concurrent Copy function is activated. Concurrent Copy requests are only accepted, however, if they can be processed without delay as a result of tape accesses:
If no tape processing times are defined, there are no restrictions.
If tape processing times are defined, Concurrent Copy requests are only accepted within the period in which tape accesses are possible.
All work files required for this function will be deleted prior to termination. If they cannot be deleted immediately for any reason (e.g. due to a system crash), they will be deleted automatically during IMPORT-PUBSET processing.
With shared pubsets, work files with the system ID of the host on which IMPORT-PUBSET is executed are deleted.
In shared pubset environment the backup job with CONCURRENT-COPY=*YES is always handled either on the local system or on the master system. The backup server settings are ignored if applicable.
WORK-FILE-NAME =
The user can specify whether the Concurrent Copy function is to be used to back up constantly changing resources (e.g. applications), to back up a pubset copy created using SHC-OSD, or to back up a Snapset. In the former case, WORK-FILE-NAME specifies the name of the work file that is to be created. This work file is used during processing to save the original contents of any files modified while backup is being performed.
If the backup process is sent to the master of a shared pubset, the work file must be accessible from the master and the slave.
WORK-FILE-NAME = *STD
The work file is assigned the following standard name:
S.DCH.<sysid>.<CC.session-id>.DATA
The work file is created on the user’s default pubset (the default pubset is defined in the user catalog).
If the ENVIRONMENT operand is set to *SYSTEM-MANAGED(...), the work file is not created on the SM pubset.
WORK-FILE-NAME = <filename 1..54 without-gen-vers>
The work file is assigned the specified name. It is not created until backup is initiated. The work file should preferably be created on an accessible pubset with, for example, a very powerful cache system.
If an existing file is to be used as the work file, its original contents are overwritten by the Concurrent Copy processing.
When files of a shared pubset are backed up, the work file must be on a pubset that the master managing the request can access.
When files of an SM pubset are backed up, the work file should be created on the
SM pubset itself. A volume set that possesses the correct attributes can be selected prior to the HSMS statement BACKUP-FILES using a CREATE-FILE command.
A further work file is created in addition to the work file containing the modified data. The additional one contains meta data and has an internally selected name. Both work files are deleted before the end of the save run. If the files cannot be deleted, e.g. because of a system failure, they will be deleted automatically later when importing.
WORK-FILE-NAME = *BY-ADDITIONAL-UNIT(...)
This operand value is only available to the HSMS administrator.
This specification applies only for backing up a pubset copy which was created in a disk storage system using SHC-OSD. To perform the backup, mirroring of the pubset volumes is interrupted and the assigned units (additional mirror or clone units, depending on the mirroring function) are detached.
The complete SF or SM pubset is taken into account. After a pubset crash it is possible to reconstruct the complete SF or SM pubset from the mirrored data volumes of the original pubset and the backed up files.
In shared-pubset operation the pubset copy is always backed up locally as the catalog accesses take place locally on the mirror units, and moving them to the master side would not enhance the performance of the catalog accesses.
DISCARD-COPY = *YES / *NO
Specifies whether the pubset copy is to be discarded after the backup. If the pubset copy is discarded, its units are available for another backup: mirroring is continued with these units, or in the case of clone units the assignment to the original disks is retained.
WORK-FILE-NAME = *BY-TARGET-CONTROLLER(...)
This operand value is only available to the HSMS administrator.
In the case of remote mirroring the target controller’s mirror units are used for saving. You disconnect the mirror units from the original volumes on the target controller in the same manner as the source controller’s mirror units (WORK-FILE-NAME=*BY-ADDITIONAL-UNIT).
In a shared pubset environment the backup job with operand value *BY-TARGET-CONTROLLER is always handled locally, i.e. it is not sent to the master or a backup server if applicable.
SELECT = *BY-ADDITIONAL-UNIT
The target controller’s mirror units are used as redundant volumes for the save.
WORK-FILE-NAME = *FROM-SNAPSET(...)
Files and job variables are transferred to a backup archive from a disk copy of the pubset which was created as a Snapset. A Snapset is identified unambiguously by the pubset specification and a letter identifier/the historical sequence number. When a backup takes place from a Snapset a new save version is generated with the creation date of the Snapset as the save time of the files.
If a newer save version is contained in the save file to be updated, SAVE-FILE=*NEW must be specified.
In a shared pubset environment the backup job with operand value *FROM-SNAPSET is always handled on the master system regardless of the BACKUP-SERVER-USAGE parameter.
The following restrictions apply for a backup from a Snapset:
Only the value *S0 is permitted in the SAVE-DATA operand. *STD is possible if S0 is preset in the archive attribute.
Only the value *PUBLIC-DISK(STORAGE-TYPE= *ANY) or *PUBLIC-DISK(STORAGE-TYPE=*PUBLIC-SPACE) is permitted in the SUPPORT operand.
Only the default value *NO is permitted in the PARTIAL-FILES operand (no partial backup in the event of a transfer from Snapsets).
Only the *ALL-FILES=*S0 suboperand (default value) is permitted for the operand value SELECT-FILES=*ALL-FILES (full backup).
The DELETE-FILES-AND-JV operand is only permitted with the default value *NO (no transfer from the Snapset with deletion on the current pubset).
PUBSET-ID = <cat-id>
Catalog ID of the pubset.SNAPSET = *LATEST / <name 1..1 with-low> / <integer -52..-1>
The copy of the pubset is identified either by an associated letter (case-sensitive) or by the historical sequence number (here -1 corresponds to the latest Snapset). The latest Snapset is preset with *LATEST (corresponds to -1).
OPERATION-CONTROL =
Enables the user to define certain parameters which are relevant to the execution of the backup run.
OPERATION-CONTROL = *STD
The default values of the operands described below apply.
OPERATION-CONTROL = *PARAMETERS(...)
The operands controlling the backup run can be modified as follows:
REQUEST-NAME = *STD / <name 1..8>
Request name that can be used in the HSMS request management statements (DELETE-REQUESTS, RESTART-REQUESTS and SHOW-REQUESTS) to refer to this request. The name is extended internally by a prefix derived from the user ID (or SYSHSMS for the HSMS administrator) and a suffix in the form of a time stamp. Unless otherwise specified, the request name is formed by the short code “BCF#” and the TSN of the calling user task yyyy as follows: BCF#yyyy.
REQUEST-DESCRIPTOR = *NONE / <text 1..60>
It is possible to enter any text that describes the request in more detail.
This text is displayed at the operator console when the request is started. The text can be output using the HSMS statement SHOW-REQUESTS.
CONTROL-JV = *NONE / <filename 1..80 without-gen-vers>
Specifies the name of a job variable that HSMS supplies with various values corresponding to important actions performed by HSMS/ARCHIVE during a complex processing operation, for this version of HSMS this only applies to complex operations, for example the processing of shadow archives.
The nonprivileged caller must be owner or co-owner of this job variable. The user issuing the request can query the job variable to obtain an overview of the current processing status. How to use the job variable is described in detail in the “HSMS Vol. 1” manual [1] in the section “Job variable for request monitoring”.
EXPRESS-REQUEST = *NO / *YES
This operand is only available to the HSMS administrator.
Determines whether tape access is to take place during the sessions defined for express requests.
WAIT-FOR-COMPLETION = *NO / *YES
Specifies whether the user wishes to wait until processing of his or her request has been completed (synchronous processing) or whether control is to be returned to the user after the validity of the HSMS statement has been checked (asynchronous processing). The maximum wait times for batch tasks and interactive tasks are different and are determined by preset global HSMS parameters.
Interactive tasks are permitted to carry out synchronous statement processing during tape sessions only.
PARALLEL-RUNS = *STD / <integer 1..16>
This operand is only significant if you are working at storage level S2. It specifies the number of save tasks (ARCHIVE subtasks) running in parallel.
PARALLEL-RUNS = *STD
The preset value from the archive definition applies if a new save file is created.
PARALLEL-RUNS = <integer 1..16>
Number of save tasks (ARCHIVE subtasks) running simultaneously.
There must be one tape device available for each task.
In the case of automatic duplication to a shadow archive, two tape devices must be available for each task.
WRITE-CHECKPOINTS = *STD / *YES / *NO
Defines whether any checkpoints are to be written to the ARCHIVE checkpoint file during processing; these checkpoints permit a request to be restarted following an interruption (state INTERRUPTED).
Unless otherwise specified, the preset value from the archive definition applies.
SHADOW-COPY =
Specifies whether the save version is to be automatically duplicated to a shadow archive which may have been assigned.
SHADOW-COPY = *ALLOWED
If the files and job variables on storage level S2 are backed up and the relevant backup archive has been assigned a shadow archive, the save version is automatically duplicated to this shadow archive after the backup run.
Whether a save file is updated in the shadow archive, a new one is created like in the main archive or a new one is always created is specified by the SHADOW-CONTROL archive attribute of the main archive.
SHADOW-COPY = *INHIBITED
The save version is not automatically duplicated to a shadow archive which may have been assigned.
TAPE-CONTROL =
Defines the parameters which are relevant for backup to tape.
TAPE-CONTROL = *STD
The preset values from the definition of the archive to be used for backup apply.
TAPE-CONTROL = *PARAMETERS(...)
The operands relevant for backup to tape can be modified as follows:
BLOCKING-FACTOR = *STD / <integer 2..15 2Kbyte > / *MAX
Blocking factor to be used for writing the save file to tape.
The default value *STD selects the blocking factor from the archive definition.
If this also contains the default value, the default values of the ARCHIVE parameters apply (SYSPAR.ARCHIVE.vvv: BLOCK-SIZE-TAPE for tapes, BLOCK-SIZE-T-C for magnetic tape cartridges).
For performance reasons it is recommmended to write blocks with maximum size to tape.
This will be achieved already by using the default values. In the case of magnetic tape cartridges this is a blocking factor of 128 (256k per block)
The values 2 to 15 should not be used anymore and are supported for compatibiliy reasons only.
*MAX selects the maximum blocking factor possible in the current BS2000 version independent of the settings in the archive and the ARCHIVE parameters. Currently this is the value 128.
UNLOAD-TAPE = *STD / *NO / *YES
Specifies whether or not tapes are to be unloaded after processing.
The preset value from the archive definition applies unless otherwise specified.
It is possible that HSMS will ignore the value of this operand in order to avoid irrational tape processing, for example to avoid unloading a tape several times while rewinding it.
DISK-CONTROL =
Defines the parameters which are relevant for backup to disk.
DISK-CONTROL = *STD
The preset values from the definition of the archive to be used for backup apply.
DISK-CONTROL = *PARAMETERS(...)
The operands relevant for backup to disk can be modified as follows:
PRIMARY-ALLOCATION = *STD / <integer 36..50331645 2Kbyte >
Size in PAM pages of the primary allocation for save file creation on disk (see the description of the CREATE-FILE command in the “Commands” manual [5]). Unless otherwise specified, the preset value from the archive definition applies.
SECONDARY-ALLOCATION = *STD / <integer 0..32767 2Kbyte >
Size in PAM pages of the secondary allocation for save file extension on disk (see the description of the CREATE-FILE command in the “Commands” manual [5]).
Unless otherwise specified (*STD), the preset value from the archive definition applies.
Values from 1 through 35 are not permitted.
WRITE-CHECK = *STD / *NO / *YES
Specifies whether a read-after-write check is to be performed after each write operation for error detection purposes. Note that read-after-write checking will result in longer execution times.
Unless otherwise specified, the preset value from the archive definition applies.
DESTROY-BY-DELETE = *STD / *NO / *YES
Specifies whether the storage space released by save file deletion is to be erased by overwriting it with binary zeros for data privacy reasons. Note that erasure by overwriting will result in longer execution times for save file deletion.
Unless otherwise specified, the preset value from the archive definition applies.
PERFORMANCE-ANALYSIS =
Specifies whether a statistics file is to be produced for each ARCHIVE subtask. ARCHIVE writes a new line to this statistics file:
when it opens a save file ARCHIVE.SAVE.FILE.
when it closes this save file.
whenever it begins saving or restoring a new file.
every 2 seconds while it is saving/restoring.
This gives a detailed overview of the ARCHIVE activities taking place during the save and restore operations.
PERFORMANCE-ANALYSIS = *NO
No statistics file is to be produced.
PERFORMANCE-ANALYSIS = *YES(...)
A statistics file is to be produced for each ARCHIVE subtask.
SEPARATOR = ; / <c-string 1..1>
Character that is inserted between the different fields in the statistics files. This enables programs such as EXCEL or LOTUS to differentiate the different fields.The default value is “ ; ”. This corresponds to the EXCEL default setting.
REPORT =
Serves to define whether a report is to be output for this backup request and to determine the scope of this report.
REPORT = *SUMMARY
A summary of the results of the backup request, including any error messages, is to be output.
REPORT = *SAVED-FILES
A full report is to be output, including a list of the files actually saved, i.e. files of save type CNS are not output.
REPORT = *FULL
A full report is to be output, including a list of all backed up files, i.e. including the files which were cataloged but not saved (save type CNS).
REPORT = *NONE
No report is to be output.
OUTPUT =
Specifies where the report is to be output.
OUTPUT = *STD
The output destination of the report is determined by the default value that is defined by the global HSMS parameter OUTPUT.
OUTPUT = *PRINTER
The report for this backup request is to be printed.
OUTPUT = *NONE
No report is output. Nevertheless the report is available as a pdf file via the SE manager application Backup Monitoring, if the monitoring is activated in the global HSMS parameters.
OUTPUT = *MAIL
The report for this backup request is sent as an email attachment to the address which is entered in the caller’s user entry. If it is not possible to send the report by email, it is printed out.
OUTPUT = *LIBRARY-ELEMENT(...)
The report for this backup request is edited and output to the specified PLAM library element (type P). If this is not possible, the report is printed out.
LIBRARY = <filename 1..54 without-gen-vers>
Name of the PLAM library.
ELEMENT = <composed-name 1..64 with-under>
Name of the element. A type P element is created with a version which contains the user ID plus the date and time.
OUTPUT = <filename 1..54 without-gen-vers>
The report for this backup request is be edited and written to the specified file.If the file already exists, the following applies:
if it is a non-empty SAM file, it is continued
if it is not a SAM file, the report is printed.
For nonprivileged callers the report files must be under the user’s own user ID. If not, co-ownership is a prerequisite.
Notes
Specification of the value *S1-STORAGE-LEVEL for the TO-STORAGE operand is permitted to the HSMS administrator and nonprivileged users. However, it is not possible to check whether storage is saturated or to restrict storage space for the S1 storage level. Uncontrolled creation of save files can saturate the disks assigned to the S1 storage level.
A single backup run started with a BACKUP-FILES statement can process the following:
files on a pubset which was declared as an S0 pubset for HSMS with the HSMS statement MODIFY-PUBSET-PARAMETERS
files on a pubset which was not declared for HSMS
The files on the undeclared pubset are processed in exactly the same way as the files on the S0 pubset. This means that
if a symbolic archive name is used, the archive assigned to the S0 pubset is used;
if a backup is being run at storage level S1, the S1 pubset assigned to the S0 pubset is used.
If CONCURRENT-COPY=*YES is specified, the following operands and operand values of the BACKUP-FILES statement are not supported:
FILE-NAMES=*OWN
FILE-NAMES=*ALL
SUPPORT=*PUBLIC-DISK(STORAGE-TYPE=*NET-STORAGE) / *PRIVATE-DISK(...) / *TAPE
(Although the value *ANY is accepted, it is implicitly changed to *PUBLIC-DISK. Thus files from private disks are not taken into account. Files residing on Net-Storage are skipped without any warning messages.)
DELETE-FILES-AND-JV=*YES
SELECT-FILES=*ALL-FILES(FROM=*LATEST-BACKUPS-OR-S0)
For compatibility reasons the value of the WRITE-CHECKPOINTS suboperand of the OPERATION-CONTROL operand is automatically set to *NO.
HSMS supports checkpoints for CCOPY backups which use SHC-OSD mirroring functions. This means that subsequent restarts are possible. When a restart takes place, it is necessary to specify OPERATION-CONTROL=*PAR(WRITE-CHECKPOINTS=*YES). (A restart is also possible if WRITE-CHECKPOINTS=*STD is set and WRITE-CHECKPOINTS has the value *YES in the specified archive.)
Example
The HSMS administrator saves data of user ID UID before inhibiting it
//BACKUP-FILES FILE-NAMES=:*:$uid., JV-NAMES=:*:$uid.,- ———————————— (1) // DELETE-FILES-AND-JV=*YES(PROTECTION-RESPECTED=*NONE), - ———————— (2) // SAVE-FILE=*NEW(RETENTION-PERIOD=365), - ———————————————————————— (3) // SAVE-VERSION-NAME=bcfuid, - ———————————————————————————————————— (4) // TO-STORAGE=*S2-STOR(VOLUMES=(tape01,tape02,tape03), - // DEVICE-TYPE='tape-c4'), - ———————————————————————————————————— (5) // OPERATION-CONTROL=*PAR(REQUEST-NAME=uid, - ————————————————————————— (6) // EXPRESS-REQUEST=*YES, - —————————————————————————————————————— (7) // REPORT=*FULL, OUTPUT=uid.filelist) ——————————————————————————— (8)
(1) | All files and job variables of user ID UID from all catalog IDs are saved to the default system backup archive. (As a prerequisite, the same system backup archive must have been assigned to all pubsets on which files of this user reside.) |
(2) | All files and job variables that are to be backed up in full are deleted after backup, regardless of any passwords or protecting attributes. |
(3) | The retention period of the save file is 1 year (365 days). |
(4) | The save version is given the name BCFUID. |
(5) | The files are saved to magnetic tape cartridges of the type TAPE-C4 at storage level S2 whose VSNs are TAPE01, TAPE02 and TAPE03. |
(6) | The request is given the name UID. |
(7) | The request is processed asynchronously as an express request. |
(8) | A full report is written to the file UID.FILELIST. |