dbed_ckptcreate - create a Storage Checkpoint for an Oracle database
dbed_ckptcreate -S ORACLE_SID -H ORACLE_HOME
[ -c checkpoint-name ]
[ -o [ online | offline | instant][,remove | noremove ] ]
[ -n ] [ -h ]
Veritas Storage Foundation. To determine whether this product is installed on the system, enter:
rpm -qa | grep VRTSdbed
The dbed_ckptcreate command creates a Storage Checkpoint for an Oracle database. Depending on whether or not the database is running, the online, offline, or instant option can be specified to take the appropriate steps to ensure a valid Storage Checkpoint.
Storage Checkpoints provide a frozen image of the datafiles in an Oracle database. Online and offline Storage Checkpoints do not provide frozen images for database control files or online redo log files. dbed_ckptcreate does, however, ensure that a backup control file is created when the Storage Checkpoint is created to allow for complete recovery, if required.
For instant Storage Checkpoints, frozen images for datafiles, control files, or online redo log files are provided. Only database-level recovery is allowed for instant Storage Checkpoints. Individual datafiles and tablespaces cannot be rolled back. In addition, rolling the database forward is not allowed for instant Storage Checkpoints (that is, you cannot apply archived redo logs).
Frozen database images can be used for data recovery in the event of logical data corruption in the database and for mounting and backing up frozen images. They can also be cloned and made writable for online modification (such as investigative work and debugging).
The following options are supported:
-S ORACLE_SID Specifies the name of the Oracle database for which a Storage Checkpoint will be created. -H ORACLE_HOME Specifies the ORACLE_HOME setting for the ORACLE_SID database. -o Specific options:
online | offline | instant Specifies the expected behavior at Storage Checkpoint time for the database. If online is specified, the expected behavior is to create a Storage Checkpoint by putting the database into hot-backup mode. If offline is specified, the database is expected to be down. If instant is specified, the database must be online and a Storage Checkpoint will be taken for later data recovery. The default option is online. remove | noremove Specifies the behavior of the file system when it runs out of space. By default, when a file system containing Storage Checkpoints runs out of space, the older Storage Checkpoints are removed to allow operations to proceed on the main file system. If it is critical to maintain the Storage Checkpoint even if the file system it resides on is out of space, specify the noremove option. -p ckpt_data_policy[,ckpt_metadata_policy] Assigns an allocation policy for Storage Checkpoint data. You can also optionally assign an allocation policy for Storage Checkpoint metadata. Before assigning an allocation policy, you must first create it. -n This option has no effect. It is retained for backwards compatibility. -h Shows command usage.
This example shows how to create a Storage Checkpoint while the database is online:
$ /opt/VRTS/bin/dbed_ckptcreate -S PROD -H /oracle/product/11g -o online Storage Checkpoint Checkpoint_971672042 created.
This example shows how to create a Storage Checkpoint while the database is offline:
$ /opt/VRTS/bin/dbed_ckptcreate -S PROD -H /oracle/product/11g -o offline Storage Checkpoint Checkpoint_971672044 created.
This example shows how to create an instant Storage Checkpoint:
$ /opt/VRTS/bin/dbed_ckptcreate -S PROD -H /oracle/product/11g -o instant Storage Checkpoint Checkpoint_971672045 created.
This example shows how to assign a Storage Checkpoint allocation policy to Storage Checkpoint data and metadata:
$ /opt/VRTS/bin/dbed_ckptcreate -S PROD -H /oracle/product/11g -o online -p ckpt_data_policy,ckpt_metadata_policy Storage Checkpoint Checkpoint_971672044 created.
dbed_ckptcreate must be run as the Oracle DBA user. The superuser cannot create Storage Checkpoints, since there is no guarantee that the database status can be correctly identified.
The SFDB repository maintains information about the tablespaces and datafiles that are allocated to a database and the file systems that these objects reside on. This information is updated when the dbed_update or vxsfadm -o getappdata command is executed. The information may be stale if modifications to tablespaces, datafiles, or file systems have occurred. It is important to update the SFDB repository whenever changes to your database are made. Some commands depend on this information being current.
dbed_ckptdisplay(1M), dbed_ckptmount(1M), dbed_ckptremove(1M), dbed_ckptrollback(1M), dbed_ckptumount(1M), dbed_clonedb(1M), dbed_update(1M), vxsfadm-oracle-checkpoint(1M), vxsfadm(1M)
Veritas Storage Foundation: Storage and Availability Management for Oracle Databases Guide
Veritas Storage Foundation for Oracle RAC Installation and Configuration Guide