Advanced - NetApp

Troubleshooting

Error Message: No Such Snapshot

Problem

You may get this error message when mounting a snapshot from a Vault Copy for a destination volume that has Deduplication enabled.

Solution

To resolve the issue, setup the following registry keys on the Client computer:

For LUN mount operation, create the sNETAPPCLONELUNBYSNAPTIMESTAMP registry key
For NFS mount operation, create the sNETAPPREVERTVOLUMEBYSNAPTIMESTAMP registry key

Snap creation fails with a busy LUN error

A snapshot enters a busy state in the following scenarios:

When there are LUN clones backed up in the snapshot
When snapshots are mounted manually outside of the SnapProtect backup, through the CommCell Console, or for SnapProtect backup operations such as backup copy creation

Busy snapshots may cause SnapProtect backups to go pending. If you have busy snapshots, consider the following:

Delete the snapshots in the reverse order they were created in.
Do not mount a volume and create another SnapProtect backup for the volume.
The SnapProtect backup software will detect busy snapshots and will not allow an additional snapshot to be created until there are no busy snaps. To avoid this snapshot dependency, do not manually create a snapshot of a volume while you have a snapshot mounted.

If you have a situation where the busy snapshot is no longer mounted but still shows as busy, proceed to delete all the additional snaps that were created while the busy snap was mounted. The snapshot will no longer be busy.

To avoid the snap dependency for NetApp ONTAP version 7.3, use the snapshot_clone_dependency volume option to enable the system to lock the backup of snapshot copies for the active LUN clone. You will be able to delete the base Snapshot copy without having to first delete all of the more recent backing Snapshot copies.

If you are using any other applications on the LUN, review the documentation for impacts caused by using the snapshot_clone_dependency volume option. If you delete the snapshot cloned by the LUN with the volume option enabled, you will not be able to restore the clone from one of the later snaps. The SnapProtect backup will detect if the dependency option is enabled and will allow additional snapshots to be created, even if the snapshot is mounted.

FAQ

What is the maximum number of snapshots created for each volume?

The maximum number of snapshot created for each volume is 255.

Can I create additional provisioning policies for secondary copies and use them for SnapProtect?

Additional provisioning policies for secondary copies can be added to the DataFabric Manager. Any provisioning policy with a name starting with “SnapProtect_” (case in-sensitive) will be displayed in the SnapProtect GUI for use in storage policy copies.

Can I Use a Non Root User to Configure an Array or NDMP?

You can use a Non-Root user for the Array configuration and NDMP configurations.

To configure the array with a non-root user, execute the following commands:

NetApp1> useradmin role add snapprotectrole -c ”[SnapProtect Management Role]" -a login-ndmp,login-http-admin,api-*

NetApp1> useradmin group add snapprotectgroup -c ”[SnapProtect Management Group]" –r snapprotectrole

NetApp1> useradmin user add snapprotectuser -c ”[SnapProtect Management Account]" -n ”S Admin" –g snapprotectgroup

For example:

NetApp1> useradmin role add snapprotectrole -c ”SP Mgmt Role" -a login-ndmp,login-http-admin,api-*
NetApp1> useradmin group add snapprotectgroup -c ”SP Mgmt Group" –r snapprotectrole
NetApp1> useradmin user add snapprotectuser -c ”SP Mgmt Account" -n ”S Admin" –g snapprotectgroup

The first command creates the SnapProtect Role with the proper rights; the second command adds this role to a newly created SnapProtect group; and the third command creates the user to for the Array credential field in the Calypso Control Panel.

Use the Same Credentials for NDMP

You can use the same credentials for NDMP that you use for the array.

To configure NDMP with the same credentials, enter the encrypted password in the following command, and then execute it:

NetApp1> ndmpd [password] snapprotectsuser

For example:

 NetApp1> ndmpd pswd1 snapprotectsuser

How Do I Verify that the Destination Filer Has SnapMirror Access to the Source Filer?

To ensure that SnapMirror jobs run, the destination filer must have SnapMirror access to the source filer. If the name or IP address of the SnapMirror filer appears in "/etc/snapmirror.allow", then the destination filer has access.

To add entries to "/etc/snapmirror.allow" execute the following command:

 source-filer> wrfile /etc/snapmirror.allow
destination-filer
<Ctrl+C> 

source-filer> rdfile /etc/snapmirror.allow
destination-filer

Alternatively, you can go into each of the hosts in DataFabric Manager and configure the the proper security rights for source and destination in the Fabric-Attached Storage settings.

How Does the SnapMirror Process Work?

1.	To get the full, detailed Snap Management plan so that SnapMirror/SnapVault can run, NetApp runs a conformation check during the first auxiliary copy job.	On NetApp On Calypso
2.	During each SnapMirror auxiliary copy job, SnapMirror creates a “BackupVersion," which is a new Snapshot of the entire volume that requires replication.
3.	Once a new Snapshot for SnapMirror is available, the Snapshot is listed under the volume on the Primary Filer.
4.	The new Snapshot also appears in the current SnapMirror session.
5.	On the destination filer site, the same Snapshot appears. In addition, if there were any previous SnapMirror on-demand jobs run, only the very last one remains on the target box. NetApp calls this a “Swizzling Process”. That is, the primary Fabric-Attached Storage (FAS) creates a delta set between the original snapshot used from the previous SnapMirror job and the new snapshot created for the current SnapMirror job. Only the details of the snapshot and the deltas are sent to the destination FAS, and refer to the previous snapshot, which was formerly the target SnapMirror session. The volume is updated to match what is in the primary FAS. The previous snapshot is not deleted because the entire process is a background job.

How Does Data Aging Work with SnapMirror?

Once a job has been Data-Aged or manually deleted, the job no longer appears in the SnapMirror target either. SnapVault is not affected by this process.

Before deletion on Calypso

Before deletion on SnapMirror

After deletion on Primary FAS

After deletion on Calypso

After deletion on the secondary FAS

The DataAging process does not remove this snapshot either. Only the next auxiliary copy job deletes this snapshot. After a new job runs, AuxCopy reports that there is no new data to move.

The primary FAS shows that a new snapshot is used for SnapMirror.

The secondary FAS shows the new snapshot here, but the process has not removed the volume snapshot that pertained to Job 18.

Now the job has been deleted from Calypso. This won’t affect the SnapMirror relationship until the next job runs and another auxiliary copy job runs. The latest SnapMirror snapshot that was synchronized to the secondary FAS is responsible for deleting the snapshot on the destination FAS box.

Best Practices

For any ONTAP version, it is recommended to provide the host name or IP address of a data port (e.g., e0A, e0B) during the NAS client or storage array configuration instead of using the management port (e.g., e0M). If you are using the management port, you must reassign the host name or IP address of the management port to a data port on the file server.
It is recommended that you do not use the system volume (typically vol0) for user data.
Ensure that enough disk space is available for snapshot operations and the disk space used by snapshots is monitored. The space requirement for snapshots created during SnapProtect backup can vary based on your environment. The space utilization is dependent on the number of snapshots created and the retention period defined for the snapshots.
It is recommended that you do not install the CommServe and DataFabric Manager on the same computer.
To perform snapshots and backup to tape operations in parallel for the same data, use two backupsets.
When planning for Snapshots, group LUNs according to their rate of change. For example, group LUNs with a high rate of change in the same volume and group LUNs with a low rate of change in another, separate volume. When calculating the size of volumes, use the rate of change to determine the amount of space necessary for Snapshots.
SnapProtect provides the full suite of capabilities for recovery regardless of the Volume to LUN relationship and ratio. However, a 1 to 1, Volume to LUN, relationship might simplify storage and application management. It is also best to ensure that each LUN resides within a Qtree in SnapMirror and SnapVault configurations. Thin Provisioned Volumes with LUNS make your storage provisioning and snapshot management flexible.
Disable the native NetApp schedule for snapshots on volumes that Calypso manages. This ensures snapshot retention and storage use is in line with the policies defined within Calypso. Using the default setting for internal snapshot scheduling might cause contention problems if multiple calls to create snapshots occur at the same time. To change the default setting, you can clear the Scheduled checkbox for the volume from Filer View or System Manager.
In order to use a specific initiator for accessing snapshots during SnapProtect backup set the sSNAP_UseINITIATOR registry key to the desired initiator address.