When you run a SnapProtect backup operation, the Calypso software performs the operations listed below in the following sequence of events:
This event includes the following operations:
This operation gathers information from local volume managers, multipath devices and physical disks (including partitions). For example, information on UUIDs is retrieved from a local volume manager.
This operation invokes the snapshot functions from the file server and creates the snapshot. Once created, Calypso records the snapshot in the CommServe database along with metadata, which includes information about local volume managers, multipaths and physical disks.
This event includes the following operations:
This operation invokes the file server functions to map clone devices to the proxy MediaAgent.
Based on the metadata collected during device detection and snapshot creation, Calypso recreates the storage hierarchy as follows:
When a partition number is detected, it is appended to the device name of the LUN. If no partition number is found, the full device name is used.
Provisioning Policies are optional to use. During the creation of the secondary snapshot copy, the following Provisioning Policies are automatically created and available for selection in the CommCell Console:
If you need to create your own provisioning policy, ensure it is defined with the following naming convention in the DFM server:
SnapProtect_<provisioning_policy_name>
where "SnapProtect_" is the prefix required for the provisioning policy to be available in the SnapProtect software.
During an Auxiliary Copy job, a new dataset is created or an existing dataset is modified with new data members, as explained in the following flow of events:
The DFM job is monitored and marked as completed/failed depending on the DFM job status.
The snapshots are copied to media in a sequential order. If you wish to perform an inline backup copy operation and a previously selected snapshot has not been copied to media, the current SnapProtect backup job will complete without creating the backup copy and you will need to create an offline backup copy for the current backup.
All virtual machines in a specific datastore can be backed up and restored together as follows:
Refer to the following for complete step-by-step instructions:
The above-mentioned steps can also be customized to group backups and restores of other entities, such as ESX Server, Resource Pools, etc.
Configuring a storage policy or snapshot copy with job based retention is recommended for File System and File System like Agents, and not for Database Agents.
Review the following scenarios if you are using the retention by number of jobs configuration:
To successfully browse and restore files on a NetApp file server that contain Unicode characters, it is recommended to use the UTF-8 setting for volume language. Please consult with NetApp for implications of changing volume language.
If you delete a snapshot copy, the following components in the DataFabric Manager server will be affected:
Secondary volumes/datasets are not deleted as they are stored in case you need to perform a restore operation from a deleted dataset. Secondary datasets can be removed manually, if needed.
No. Multi Instancing is not supported for SnapProtect backup.
No. Restore of SnapProtect backup data using Media Explorer is not supported. However, Media Explorer can be used to restore SnapProtect backup data from the Primary Copy.
For DB2, SAP for Oracle, and Oracle iDataAgents, when restoring from a snapshot using file system, the number of streams depends on the number of mount points used for the restore operation. Similarly, when restoring from a backup copy using file system, the number of streams depends on the number of media groups used for the restore operation.
Yes. It is supported to run a SnapProtect backup operation of Raw Device Map (RDM) disks for any Windows-based Agent configured in a virtual machine. Ensure the iSCSI Initiator is configured on the client computer to access the storage device in order to perform a successful SnapProtect operation.
During a SnapProtect backup job, the Calypso software runs a set of commands to discover storage devices.
The following table displays the commands that are used to list, rescan and delete adapters/devices for each operating system:
| Operating System | List FIBRE cHANNEL aDAPTERS | Rescan FIBRE cHANNEL aDAPTERS | Rescan ISCSI Devices | Delete Device |
AIX |
<Base>/scan_fc_adapters.sh | cfgmgr -l <hba_name> | cfgmgr -l <iscsi_adapter> | rmdev -d -l <hdisk_name> |
HP-UX |
<Base>/scan_fc_adapters.sh |
|
For Legacy HP-UX (earlier than 11.23): rmsf –a <device> For non-Legacy HP-UX:
|
|
Linux |
<Base>/scan_fc_adapters.sh |
|
|
echo 1 > /sys/block/<device>/device/delete |
Solaris |
<Base>/scan_fc_adapters.sh | cfgadm -c configure <controller> | devfsadm -i iscsi | devfsadm -C -c disk |
Yes, you can customize the device deletion and the rescan logic by creating the following registry keys:
| Key | Description |
| DeleteDevice | This key will trigger a custom script which will perform an OS level
device deletion. Once all the devices are removed, the Calypso software will remove the LUNs from the storage
arrays. Use this key if you need to unmount snapshots that are mapped to either the proxy or source machine. The key will delete all devices mapped to the machine. Follow the steps below to create this registry key:
|
| RefreshBus | This key will trigger a custom script for each host bus adapter that
needs to be rescanned. Use this key if you need to rescan the machine for any new devices each time a snapshot is mounted or unmounted. Follow the steps below to create this registry key:
|
After creating the above registry keys, you can use the following script examples:
When you configure an storage device using Array Management in the CommCell Console, the Device Group text box defines the target device group where Calypso will move the source and target devices during SnapProtect backup operations. This option does not specify the source device group (from where a device should be found during Snap/Clone operations).
The following table explains the default behavior for selecting a Snap or Clone target device:
| Scenarios based on Array Management Options | Target device is selected from... | Source and Target devices are moved to... |
| Device Group is not specified and Use devices only from this device group is not checked | Any device in the array | CV_PROTECTION_GROUP device group |
| Device Group is not specified but Use devices only from this device group is checked | CV_PROTECTION_GROUP device group | CV_PROTECTION_GROUP device group |
| Device Group is specified but Use devices only from this device group is not checked | Any device in the array | The device group specified in Device Group |
| Device Group is specified and Use devices only from this device group is checked | The device group specified in Device Group | The device group specified in Device Group |
Yes. You can use File System iDataAgent and perform the SnapProtect operation to backup the physical RDMs in the following scenario:
Yes, you can override the default restore selection by choosing a different Virtual Centre or ESX Server to restore or moving to tape, by following the steps given below:
Yes, the Virtual Server client can support advanced transport modes like SAN during restores and mounting operations, if configured as follows:
If a virtual machine undergoing a backup job includes independent disks, physical or virtual RDMs, these disks will be skipped. During a full VM restore the independent disk/Physical or virtual RDMs will get restored as a regular disk with 0MB data.
If a subclient contains virtual machines with independent disks/physical or virtual RDMs, the backup job will always complete with the status "Completed w/ one or more errors". However, if you create the IgnoreUnsupportedDisks registry key on the proxy computer, the backup job will complete successfully.
Virtual RDMs are protected by the backup job (but not during IntelliSnap backup). However at the time of restore, the data is restored as a regular VMDK on a datastore. A virtual RDM is not re-created and the data is not restored to the virtual RDM’s device.
Yes. To successfully run a SnapProtect backup of a virtual machine with Windows 2008 R2, ensure that if the virtual machine resides on a single datastore it is not spread across multiple folders.
If the virtual machine is spanned across multiple datastores, you can run SnapProtect backups for virtual machines with ESX version 4.1 (or higher) by configuring the VM using the steps below:
No, data that resides on the root volume cannot be part of the virtual machine content.
No. The archive log destination should be a valid directory for snap. Snap protect operation will not work if you configure the archive log destination with a different pattern such as log_Archive_dest_1='location=/archivelog/db_'.
For VM backups, capacity licensing is based on the total backup size, calculated as the sum of backup sizes for all VM backup jobs after white spaces (blocks of zeros) are removed. The license counts the backup size of all configured subclients; virtual machines that are included in multiple subclients will be counted multiple times. The backup size is measured for usage tracking and shown on the Backup Job Summary Report.
The backup size can be different from the guest host size or used space value shown for the VM in the disk properties dialog by Microsoft Windows.
The following factors can affect the backup size calculation:
For example:
The backup size reflects the size after eliminating white spaces; but data that was written and deleted still counts as reserved (allocated) space. The layering effects of multiple virtual file systems can cause differences between the size reported by the guest host running within the VM and the reported backup size. Frequent deletion of large files can easily cause these numbers to be out of sync.
Version 9.0 reports on all allocated blocks in the VM. The amount reported for allocated blocks can be the same size or larger than what is actually in use and can contained reserved space for deleted items. For each VMware instance, Version 9.0 has an additional reporting column of the actual size of VMs.
The following measures can help reduce backup size:
For VM backups, capacity licensing is based on the total backup size, calculated as the sum of backup sizes for all VM backup jobs after white spaces (blocks of zeros) are removed. The license counts the backup size of all configured subclients; virtual machines that are included in multiple subclients will be counted multiple times. The backup size is measured for usage tracking and shown on the Backup Job Summary Report.
The backup size can be different from the guest host size or used space value shown for the VM in the disk properties dialog by Microsoft Windows.
The following factors can affect the backup size calculation:
For example:
The backup size reflects the size after eliminating white spaces; but data that was written and deleted still counts as reserved (allocated) space. The layering effects of multiple virtual file systems can cause differences between the size reported by the guest host running within the VM and the reported backup size. Frequent deletion of large files can easily cause these numbers to be out of sync.
Version 9.0 reports on all allocated blocks in the VM. The amount reported for allocated blocks can be the same size or larger than what is actually in use and can contained reserved space for deleted items. For each VMware instance, Version 9.0 has an additional reporting column of the actual size of VMs.
The following measures can help reduce backup size:
Yes, you can perform SnapProtect operations for a single node Oracle RAC setup. The following configurations are required:
No. The Use Separate Proxy for Snap to Tape option in Subclient Properties is not supported for the Oracle iDataAgent.
Generally, we can set copy precedence for storage policy. The copy that is set to a copy precedence of 1 will be restored. However, if a snapprotect backup job is available, then it won't consider the copy precedence and restores only from the snapprotect backup.
Yes. You can perform an incremental backup copy using RMAN.
No. The table level restores will not work, if the database and instance names are different.
No. You cannot prevent a single volume revert when the database is spanned across multiple volumes. When you perform a hard revert, the database will not be restored as the reversion happens at LUN level and the other source paths related to this database are not reverted. Hence, we recommend Restore by Revert which is application aware instead of regular revert using the CommCell Console (GUI).
The following steps explain the current process of reversion for a single volume when the database is spanned across multiple volumes:
No. The data that resides on the root volume cannot be part of the Oracle database content.
No. Log Only restore is not supported for ASM database.
No. Softlinks are not supported for this iDataAgent's datafile paths and archive log location. you should use the real paths.
You can create alias device using mknod for raw devices.
The following table displays the supported IntelliSnap operations based on the location of the Oracle Application, Data and Log volumes on the client :
| Operation | Oracle APPlication / data / log are on the same Physical disk/Volume Group | Oracle APP is on different device / Volume Group but Oracle data and log on Same Physical disk / Volume Group | Oracle APPlication, data and log are on different Physical disks / Volume Groups (Recommended Configuration) | Notes |
| IntelliSnap Backup | Supported | Supported | Supported | |
| Backup Copy | Supported | Supported | Supported | |
| IntelliSnap Restore | Supported | Supported | Supported | |
| Volume level Revert | Not Supported | Supported (based on Array Support) | Supported | For NetApp array, if data and log are on the same volume, log snap will be deleted automatically after reverting the data volume. |
| File Level Revert (NetApp NFS only) | Supported | Supported | Supported |
If data and logs are on same volume/disk, multiple snaps will be taken in data phase and log phase separately.
If the SnapProtect backup and Clone exist for the same source device in the case of EMC Symmetrix, the revert operation from SnapProtect backup will work. However, the revert from clone will not work until all the snap sessions are terminated.
No. Currently, the Log only SnapProtect backup is not supported for multiple instances using a shared storage on a client
No. Currently, the table level and ASM SnapProtect backups are not supported in a multiple instances environment using a shared storage on a client.
Yes. If the data you want to backup resides on a vFiler, set it up as follows:
If you plan to add more vFilers as NAS clients, ensure they have unique names. vFilers with the same name are not supported.
No, data that resides on the root volume cannot be part of the NAS subclient content.
No. vFilers cannot be the destination location for SnapVault/SnapMirror copies as it is not supported by the DataFabric Manager.
No. Restore of NAS data to a Windows or Unix computer using File System NDMP Restore Enabler (NRE) is not supported for NAS iDataAgent snapshots.
Yes, you can enable the Image Backup Set option (for SnapMirror to Tape) on a backup set containing the SnapVault copy. However, the image backup of the SnapVault copy will not exactly match the primary (source) volume. For example, the non-qtree data on the primary volume will be in a qtree on the SnapVault copy; the SnapVault copy may also contain data from other primary volumes.
No. Restores from NAS Snapshots are not supported. You can restore to a client from a NAS backup copy.
The maximum number of snapshots created for a USP volume is 3. When using the HDS storage array with SAP for Oracle, you can have up to 4 snapshots on the same volume if you store the Data ($SAPDATA_HOME) and Config ($ORACLE_HOME/dbs) directories in the same volume.
No. Softlinks are not supported for this iDataAgent's datafile paths and archive log location. you should use the real paths.
No. During a SnapProtect backup, log files are not moved to the snapshot copy even if the Backup Log Files option is selected on the subclient where the backup operation is being performed. Log files always use the traditional backup method, not the SnapProtect backup.
Running a restore operation from the snapshot copy will fail because the log data will not be found. If you want to restore both data and logs, modify the copy precedence to restore from the Primary(Classic) copy.
No. Softlinks are not supported for this iDataAgent's datafile paths and archive log location. you should use the real paths.
No, root folder (/) should not be included as subclient content for SnapProtect.
Qlogic SANSurfer CLI on SuSE Linux Enterprise Server 10 SP2.
No, the restore of ACLs from one operating system to another is not supported.
The restore of ACLs from one file system type to another may fail as the ACLs restore is dependent on the file system implementation. For example, ACLs which are backed up from ext3 cannot be restored to NFS and vice versa. Also, if you attempt to restore data from one file system type to another with ACLs included, but the file system is mounted without ACLs, the restore operation will fail. To workaround this issue ensure that the file system is mounted with ACLs before attempting a restore operation.
To prevent the restore operation from failing you can restore data without the ACLs included.
When you revert all the logical volumes in the AIX Logical Volume Manager, the revert operation supports all logical volume types and all their attributes are preserved. However, for revert operations on single logical volumes, striped volumes are not supported.
The following table displays the logical volume attributes that are preserved when you revert a single volume:
| Attribute Name | Description |
inter |
Inter-physical volume allocation policy |
intra |
Intra-physical volume allocation policy |
label |
Volume label |
Relocatable |
Defines if relocation is allowed during reorganization of the logical volume |
strictness |
Strict allocation policy |
type |
Logical volume type |
upperbound |
Upper-bound on the physical volume used for new allocations |
permission |
Access permission of the logical volume |
mirror write consistency |
Defines if mirrored copies are in a consistent/active state |
serialize IO |
Defines if logical volumes are overlapping I/O serialized |
sched policy |
Defines which scheduling policy is being used when more that one logical partition is written |
BB policy |
Bad block relocation policy |
write verify |
Defines if all verified write operations are followed by a follow up read operation |
No. Softlinks are not supported for SnapProtect operation. you should use the real paths.
You can create alias device using mknod for raw devices.
The incremental or differential backup will run as full backup and include all the files located on the source in the backup, even though these backups will be displayed as incremental or differential in the Job Manager.
You can enable the classic scan or Data Classification to scan files during a SnapProtect backup. If you select the Data Classification as the scanning method, ensure that the meta database is located on the same volume for which you are performing the snap backup.