Use this command to generate a backup set for a Backup-Archive
Client node. A backup set is a collection of a Backup-Archive
Client's active backed up data, which is stored and managed as a single
object, on specific media, in server storage. Although you can create
a backup set for any client node, a backup set can be used only by
a Backup-Archive Client.
Restriction: A backup set in
"deduplication
format" has that designation as a result of a
GENERATE
BACKUPSET command with at least one of the following specifications:
- Includes a node at Backup-Archive Client level 6.1.x (at least
6.1.0 but less than 6.2.0).
- Includes a node that has one or mode nodes that are authorized
to act as a proxy. At least one of those proxy nodes is at Backup-Archive
Client level 6.1.x.
Backup sets in the deduplication format can be restored only
by the 6.1.2 or later Backup-Archive Client. Backup-Archive Clients
before version 6.1.2 cannot restore from a backup set that is in the
deduplication format.
A backup set in the
"distributed deduplication
format" has that designation as a result of a
GENERATE
BACKUPSET command with at least one of the following specifications:
- Includes a node at Backup-Archive Client level 6.2.0 or later.
- Includes a node that has one or mode nodes that are authorized
to act as a proxy. At least one of those proxy nodes is at Backup-Archive
Client level 6.2.0.
Backup sets in the distributed deduplication format can be restored
only by the 6.2.0 or later Backup-Archive Client.
Restriction: You cannot generate a backup set with files that
were backed up to Tivoli® Storage
Manager using
NDMP. However, you can create a backup set with files that were backed
up using NetApp SnapShot Difference.
The server creates copies
of active versions of a client's backed up objects that are within
the one-or-more file spaces specified with this command. The server
then consolidates them onto sequential media. Currently, the backup
object types that are supported for backup sets include directories
and files only.
The backup-archive client node can restore its
backup set from the server and from the media to which the backup
set was written.
This command generates a background process
that can be canceled with the CANCEL PROCESS command.
If the background process created by this command is canceled, the
media might not contain a complete backup set. You can use the QUERY
PROCESS command to show information about the background
process that is created by this command.
Tip: When Tivoli Storage
Manager generates
a backup set, you can improve performance if the primary storage pools containing
the client data are collocated. If a primary storage pool is collocated,
client node data is likely to be on fewer tape volumes than it would
be if the storage pool were not collocated. With collocation, less
time is spent searching database entries, and fewer mount operations
are required.
Privilege class
To issue this command, you
must have system privilege or policy privilege for the domain to which
the client node is assigned.
Syntax
.-,-------------------.
V |
>>-GENerate BACKUPSET----+-node__name------+-+------------------>
'-node_group_name-'
.-*-------------------.
>--backup_set_name_prefix--+---------------------+-------------->
| .-,---------------. |
| V | |
'---file_space_name-+-'
.-SCRatch--=--Yes-----.
>--DEVclass--=--device_class_name--+---------------------+------>
'-SCRatch--=--+-Yes-+-'
'-No--'
>--+------------------------------+----------------------------->
| .-,------------. |
| V | |
'-VOLumes--=----volume_names-+-'
.-RETention--=--365---------.
>--+---------------------------+-------------------------------->
'-RETention--=--+-days----+-'
'-NOLimit-'
.-Wait--=--No------.
>--+-----------------------------+--+------------------+-------->
'-DESCription--=--description-' '-Wait--=--+-No--+-'
'-Yes-'
.-NAMEType--=--SERVER------.
>--+--------------------------+--------------------------------->
'-NAMEType--=--+-SERVER--+-'
+-UNIcode-+
'-FSID----'
.-CODEType--=--BOTH-----------.
>--+-----------------------------+------------------------------>
'-CODEType--=--+-UNIcode----+-'
+-NONUNIcode-+
'-BOTH-------'
.-PITDate--=--current_date-. .-PITTime--=--current_time-.
>--+--------------------------+--+--------------------------+--->
'-PITDate--=--date---------' '-PITTime--=--time---------'
.-DATAType--=--FILE----------. .-TOC--=--Preferred-----.
>--+----------------------------+--+-----------------------+---->
| .-,---------. | '-TOC--=--+-No--------+-'
| V | | +-Preferred-+
'-DATAType--=----+-FILE--+-+-' '-Yes-------'
+-IMAGE-+
'-ALL---'
>--+-----------------------------+------------------------------>
'-TOCMGmtclass--=--class_name-'
.-ALLOWSHREDdable--=--No------.
>--+-----------------------------+-----------------------------><
'-ALLOWSHREDdable--=--+-No--+-'
'-Yes-'
Parameters
- node_name or node_group_name (Required)
- Specifies the name of the client node and node groups whose data
is contained in the backup set. To specify multiple node names and
node group names, separate the names with commas and no intervening
spaces. You can use wildcard characters with node names but not with
node group names. When multiple node names are specified, the server
generates a backup set for each node and places all of the backup
sets together on a single set of output volumes.
- backup_set_name_prefix (Required)
- Specifies the name of the backup set for the client node. The
maximum length of the name is 30 characters.
When you select a
name, Tivoli Storage
Manager adds
a suffix to construct your backup set name. For example, if you name
your backup set mybackupset, Tivoli Storage
Manager adds a unique
number such as 3099 to the name. The backup set name is then identified
to Tivoli Storage
Manager as mybackupset.3099.
To later show information about this backup set, you can include a
wildcard with the name, such as mybackupset.* or
specify the fully qualified name, such as mybackupset.3099.
When
multiple node names or node group names are specified, the server
generates a backup set for each node or node group and places all
the backup sets on a single set of output volumes. Each backup set
is given the same fully qualified name consisting of the backup_set_name_prefix and
a suffix determined by the server.
- file_space_name
- Specifies the names of one or more file spaces that contain the
data to be included in the backup set. This parameter is optional.
The file space name that you specify can contain wildcard characters.
You can specify more than one file space by separating the names with
commas and no intervening spaces. If you do not specify a file space,
data from all the client nodes backed-up and active file spaces is
included in the backup set.
For a server that has clients with support
for Unicode-enabled file spaces, you can enter either a file space
name or a file space ID (FSID). If you enter a file space name, you might need to
have the server convert the file space name that you enter. For example,
you might need to have the server convert the name that you enter
from the server's code page to Unicode. See the NAMETYPE parameter
for details. If you do not specify a file space name, or specify only
a single wildcard character for the name, you can use the CODETYPE
parameter to limit the operation to Unicode file spaces or to non-Unicode
file spaces.
- DEVclass (Required)
- Specifies the name of the device class for the volumes to which
the backup set is written. The maximum length of the name is 30 characters.
Restriction: You cannot specify a device class with a device
type of NAS or CENTERA.
- SCRatch
- Specifies whether to use scratch volumes for the backup set. If
you include a list of volumes using the VOLUMES parameter, the server
uses scratch volumes only if the data cannot be contained in the volumes
you specify. The default is SCRATCH=YES. The values are:
- YES
- Specifies to use scratch volumes for the backup set.
- NO
- Specifies not to use scratch volumes for the backup set.
- VOLumes
- Specifies the names of one or more volumes that will contain the
backup set. This parameter is optional. You can specify more than
one volume by separating each volume with a comma, with no intervening
spaces.
If you do not specify this parameter, scratch volumes are
used for the backup set.
- RETention
- Specifies the number of days to retain the backup set on the server.
You can specify an integer from 0 to 30000. The default is 365 days.
The values are:
- days
- Specifies the number of days to retain the backup set on the server.
- NOLimit
- Specifies that the backup set should be retained on the server
indefinitely.
If you specify NOLIMIT, the server
retains the volumes containing the backup set forever, unless a user
or administrator deletes the volumes from server storage.
- DESCription
- Specifies the description to associate with the backup set. This
parameter is optional. The maximum length of the description is 255
characters. Enclose the description in quotation marks if it contains
any blank characters.
- Wait
- Specifies whether to wait for the server to complete processing
this command in the foreground. This parameter is optional. The default
is NO. The values are:
- Yes
- Specifies the command processes in the foreground. Messages that
are created are not displayed until the command completes processing.
You cannot specify WAIT=YES from the server console.
- No
- Specifies that the command processes in the background. Use the QUERY
PROCESS command to monitor the background processing of
this command.
- NAMEType
- Specify how you want the server to interpret the file space names
that you enter. This parameter is useful when the server has clients
with support for Unicode-enabled file spaces. You can use this parameter
for Tivoli Storage
Manager clients using Windows, NetWare, or Macintosh OS X operating
systems.
Use this parameter only when you enter a partly or fully
qualified file space name. The default value is SERVER. Possible values are:
- SERVER
- The server uses the server's code page to interpret the file space
names.
- UNIcode
- The server converts the file space name that is entered from the
server code page to the UTF-8 code page. The success of the conversion
depends on the actual characters in the name and the server's code
page. Conversion can fail if the string includes characters that are
not available in the server code page, or if the server has a problem accessing
system conversion routines.
- FSID
- The server interprets the file space names as their file space
IDs (FSIDs).
Important: Use care when specifying this parameter
if multiple node names are also specified. Different nodes might use
the same file space ID for different file spaces, or different file
space IDs for the same file space name. Therefore, specifying a file
space ID as the file space names can result in the wrong data being
written to the backup set for some nodes.
- CODEType
- Specify what type of file spaces are to be included in the operation.
The default is BOTH, meaning that file spaces are included regardless
of code page type. Use this parameter only when you enter a single
wildcard character for the file space name or when you do not specify
any file space names. Possible values are:
- UNIcode
- Include only file spaces that are in Unicode.
- NONUNIcode
- Include only file spaces that are not in Unicode.
- BOTH
- Include file spaces regardless of code page type.
- PITDate
- Specifies that files that were active on the specified date and
that are still stored on the Tivoli Storage
Manager server are
to be included in the backup set, even if they are inactive at the
time you issue the command. This parameter is optional. The default
is the date on which the GENERATE BACKUPSET command
is run. You can specify the date using one of the following values:
Value |
Description |
Example |
MM/DD/YYYY |
A specific date |
09/15/1998 |
TODAY |
The current date |
TODAY |
TODAY-daysor-days |
The current date minus days specified |
TODAY-7 or -7. To include files that were
active a week ago, specify PITDATE=TODAY-7 or PITDATE=-7
|
EOLM (End Of Last Month) |
The last day of the previous month. |
EOLM |
EOLM-days |
The last day of the previous month
minus days specified. |
EOLM-1 To include files that
were active a day before the last day of the previous month.
|
BOTM (Beginning Of This Month) |
The first day of the current month. |
BOTM |
BOTM+days |
The first day of the current month,
plus days specified. |
BOTM+9 To include files that
were active on the 10th day of the current month.
|
- PITTime
- Specifies that files that were active on the specified time and
that are still stored on the Tivoli Storage
Manager server are
to be included in the backup set, even if they are inactive at the
time you issue the command. This parameter is optional. IF a PITDate was specified, the default is
midnight (00:00:00); otherwise the default is the time at which the GENERATE
BACKUPSET command is started. You can specify the time using
one of the following values:
Value |
Description |
Example |
HH:MM:SS |
A specific time on the specified PIT date |
12:33:28 |
NOW |
The current date on the specified PIT date |
NOW |
NOW+HH:MM or +HH:MM |
The current time plus hours and minutes on the
specified PIT date |
NOW+03:00 or +03:00 If you issue this command
at 9:00 with PITTIME=NOW+03:00 or PITTIME=+03:00. Tivoli Storage
Manager includes
files that were active at 12:00 on the PIT date.
|
- DATAType
- Specifies that backup sets containing the specified types of data
that are to be generated. This parameter is optional. The default
is that file level backup sets are to be generated. To specify multiple
data types, separate data types with commas and no intervening spaces.
The
server generates a backup set for each data type and places all the
backup sets on a single set of output volumes. Each backup set is
given the same fully qualified name consisting of the backup_set_name_prefix and
a suffix determined by the server. However, each backup set has a
different data type, as shown by the QUERY BACKUPSET command.
Possible values are:
- ALL
- Specifies that backup sets for all types of data (file level,
image, and application) that have been backed up on the server are
to be generated.
- FILE
- Specifies that a file level backup set is to be generated. File
level backup sets contain files and directories that are backed up
by the backup client. If no files or directories have been backed
up by the backup client, a file level backup set is not generated. This
is the default.
- IMAGE
- Specifies that an image backup set is to be generated. Image backup
sets contain images that are created by the backup client BACKUP
IMAGE command. Image backup sets are generated only if an
image has been backed up by the backup client.
- TOC
- Specifies whether a table of contents (TOC) is saved for each
file level backup set. Tables of contents are always saved for backup sets containing
image or application data. The TOC parameter is ignored when generating
image and application backup sets. A table of contents will always be generated
for image and application backup sets.
Consider the following in
determining whether you want to save a table of contents:
- If a table of contents is saved for a backup set, you can use
the Tivoli Storage
Manager Web
backup-archive client to examine the entire file system tree and choose
files and directories to restore. To create a table of contents, you
must define the TOCDESTINATION attribute in the backup copy group
for the management class that is specified by the TOCMGMTCLASS parameter.
Creating a table of contents requires additional processing, storage
pool space, and possibly a mount point during the backup set operation.
- If a table of contents is not saved for a backup set, you can
still restore individual files or directory trees using the backup-archive
client RESTORE BACKUPSET command, if you know the
fully qualified name of each file or directory to be restored.
To display the contents of backup sets, you can also use the
QUERY
BACKUPSETCONTENTS command.
This parameter is optional.
Possible values are:
- No
- Specifies that table of contents information is not saved for
file level backup sets.
- Preferred
- Specifies that table of contents information should be saved for
file level backup sets. This is the default. However, a backup set
does not fail just because an error occurs during creation of the
table of contents.
- Yes
- Specifies that table of contents information must be saved for
each file level backup set. A backup set fails if an error occurs
during creation of the table of contents.
- TOCMGmtclass
- Specifies the name of the management class to which the table
of contents should be bound. If you do not specify a management class,
the table of contents is bound to the default management class for
the policy domain to which the node is assigned. In this case, creation
of a table of contents requires that you define the TOCDESTINATION
attribute in the backup copy group for the specified management class.
- ALLOWSHREDdable
- Specifies whether data from a storage pool that enforces shredding
is included in the backup set. This parameter is optional. Possible
values are:
- No
- Specifies that data from a storage pool that enforces shredding
is not included in the backup set. This is the default.
- Yes
- Specifies that data from a storage pool that enforces shredding
can be included in the backup set. The data on the backup set media will
not be shredded.
Example: Generate a backup set for a file space
Generate
a backup set of a file space that is called
/srvr that
belongs to client node JANE. Name the backup set PERS_DATA and retain
it for 75 days. Specify that volumes VOL1 and VOL2 contain the data
for the backup set. The volumes are to be read by a device that is
assigned to the AGADM device class. Include a description.
generate backupset jane pers_data /srvr devclass=agadm
retention=75 volumes=vol1,vol2
description="area 51 base image"
Example: Generate a backup set of a Unicode-enabled
file space
Generate a backup set of the Unicode-enabled file
space,
\\joe\c$, that belongs to client node
JOE. Name the backup set JOES_DATA. Specify that volume VOL1 contain
the data for the backup set. The volume is to be read by a device
that is assigned to the AGADM device class. Have the server convert
the
\\joe\c$ file space name from the server
code page to the UTF-8 code page.
generate backupset joe joes_data \\joe\c$ devclass=agadm
volumes=vol1 nametype=unicode