Skip to main content

We've Moved!

Product Documentation has moved to docs.hitachivantara.com
Hitachi Vantara Knowledge

Data Protection Workflows Overview

Each data protection workflow described here is a high level task that references of a series of subtasks.

Subtasks are detailed, lower level descriptions that explain how to do something via the UI or CLI.

Workflows and subtasks both refer out to user interface reference topics that describe the details of a particular web page, wizard or dialog that the user interacts with.

When following a workflow for the first time, it is recommended that you study the related subtasks and reference topics before carrying out the steps. This will ensure that you are aware of any notes, warnings and cautions that may apply.

Once you are familiar with the workflows and subtasks, it may only be necessary to refer to the reference topics, since all the important notes, warnings and cautions are located within the reference topics at the relevant point.

TipRefer to Button Icons to find the icons associated with the control names used in the step-by-step guides.

How to protect your data

Before you begin

This generic workflow describes the steps for protecting your data with Ops Center Protector.

It is assumed that the following tasks have been performed:

This task describes the general steps to follow when protecting your data:

Procedure

  1. Create the required source nodes and then check that they are authorized and online.

    Source nodes represent the places where the data to be protected resides in your system. See How to add a node and How to authorize a node.
    NoteSource Hardware Storage Device nodes will need to be created even if they don't appear on the data flows for snapshot and replication operations.
  2. Create the required destination nodes and then check that they are authorized and online.

    Destination nodes represent the places where the data is to be backed up to. See How to add a node and How to authorize a node.
  3. Define the data protection policies.

    Policies define the data sets to be protected and the methods to be used to protect them. See How to create a policy.
  4. Draw the backup data flows.

    Data flow diagrams show the participating source and destination nodes and the data paths interconnecting them. See How to create a data flow.
  5. Assign the policies to the participating nodes on the data flows.

    Policy assignments define precisely how the data flows from each source node to the respective destination nodes. See How to apply a policy to nodes on a data flow.
  6. Compile and activate the data flows.

    The source and destination nodes work autonomously by implementing rules locally. These rules are generated by the master node and disseminated to the participating nodes. See How to activate a data flow.
  7. Trigger the policies if required.

    Policies are invoked according to a user defined schedule and/or RPO. In some cases it may be necessary or desirable to trigger policies manually. See How to create a schedule, How to trigger an operation from an active data flow and Triggering policies and operations with hdidcmd.
  8. Monitor the data flows, logs etc. to ensure policies are operating as expected.

    A number of other tools are provided that enable the correct functioning of data protection policies to be confirmed, including the:
  9. Review the storage nodes to ensure backups are being created.

    The various storage devices require ongoing surveillance to ensure that they are operating correctly and sufficient resources are available to store your data securely. These are shown in the Storage Inventory.

How to restore your data

Before you begin

This generic workflow describes the steps for restoring data that is protected with Ops Center Protector.

It is assumed that a data protection policy has been implemented and that at least one complete backup data set has been created on the designated storage device. See How to protect your data.

This task describes the general steps to follow when restoring your data:

Procedure

  1. Identify the destination where the data set is to be restored.

    Depending on the scenario, you may want to restore data to its original location or a different location
  2. Ensure that the restore location is prepared to receive the restored data set. In the case of a replication, it will be necessary to prepare both sides of the replication pair.

    Applications that access the data will need to be placed in the correct state and sufficient resources must be available to receive the data set.
    CautionThe process of restoring data can result in the destruction of some original data that exists on the restore target. Ensure that any critical data is copied to a safe location and/or is included in the data set being restored.
  3. Depending on the scenario, it may be necessary to suspend any backup policies currently active on the location where the data set is being restored.

    Data flows can be deactivated via the Monitor Inventory or Data Flows Inventory.
  4. Locate the data set to be restored.

    The Restore Inventory and Storage Inventory tracks all backups maintained by Protector.
  5. Select the restore method and start the restore process.

    It may be possible to restore the data to the original location, a different location, mount a volume to a new host or swap the direction of a replication. The Restore Inventory and Storage Inventory provide functions to restore, mount, revert and swap replications and snapshots.
  6. Once the restore process is complete, further steps may be needed to fix-up the data set before using it.

    For supported applications, these additional steps are described in the appropriate Application Guide (see Related documents). For other applications, consult the vendor's documentation.
  7. Restart any applications that access the restored data.

    For supported applications, these additional steps are described in the appropriate Application Guide (see Related documents). For other applications, consult the vendor's documentation.
  8. Resume any existing backup policies for the restored data set. If a new restore location was chosen, consider if it is necessary to introduce new backup policies.

    Data flows can be reactivated via the Data Flows Inventory.

How to protect Ops Center Protector

Before you begin

In order for Ops Center Protector Master Node backups to be available following the total loss of the master node backups should be made to remote storage. Such storage may be a repository that is proxied by a client node HCP; HCP for cloud scale or Amazon S3.

When using HCP; HCP for cloud scale or Amazon S3 for backup either a client node or the master itself may be the proxy for the storage node.

When using a repository that repository must be hosted (i.e. proxied) by a client node. Note that only generation 2 repositories should be used.

Backing up the Protector Master Node Settings

Procedure

  1. Create a policy. Refer How to create a policy

  2. Add classification of type "Ops Center". Refer How to add a classification to a policy

  3. For that classification select the master node, then under "Include Applications" select either "Include all Ops Center applications" or press the Add button and select "Ops Center Protector Master". Refer Ops Center Classification Wizard .

  4. Add a backup operation to the policy with suitable schedule and/or RPO, etc

  5. Create a dataflow. Refer How to create a data flow .

  6. Drag the master node onto the data flow, the master node will be the source node.

  7. Select the policy created above on that master. Refer How to apply a policy to nodes on a data flow.

  8. Drag the target storage node onto the data flow and connect the master source node to the repository. Refer How to connect nodes on a data flow.

  9. Select the backup operation in the above policy on the storage node. Refer How to apply a policy to nodes on a data flow.

  10. Activate the dataflow. Refer How to activate a data flow.

    This complete the backup of the protector settings, they will be automatically backed up in accordance with the schedule and/or RPO selected within the policy.

    An additional backup may be performed by selecting the data flow from within the "Monitor" screen. Refer How to trigger an operation from an active data flow.

How to recover Ops Center Protector

Before you begin

Ops Center Protector master node backups will be required when attempting to rebuild that master node whether that node is to be hosted on the original hardware or on a new system

When recovering to a new system that system must not be in used as a Ops Center Protector.

Where cloud storage has been used for backups this new system must have an internet connection and be able to communicate with required cloud storage system.

Where a repository hosted (proxied) by a client node has been used for backups you will need to know the directory path for that repository. This is the directory path that was specified when the repository was created.

In the event that the Master node is lost or otherwise permanently unavailable all settings for that master node may be recovered to a new master as described below. Note that if testing this procedure while the original master node is still available that master MUST be stopped prior to the commencement of the following actions.

Procedure

  1. Create a new Master Node

    • Install Protector as a recovery master node on a new system (see How to install/upgrade Protector on Windows and Linux or AIX . Where the system has previously had Ops Center Protector installed ensure it is completely uninstalled before re-installing (i.e. ensure the install directory has been removed)
    • Give the new master a temporary name (e.g. Recovery_Master) This will be overwritten later in this procedure with the original master node name.
    • Note the IP address of the new master.
  2. Preparing a Repository Storage Proxy Node for use by the new Master Node

    NoteThis section can be ignored if the Ops Center Protector Master Node backups are not stored on a repository node:

    On the client node that is proxy for the repository containing the protector settings back ups:

    • Stop the protector services. Refer How to stop the services.
    • To prepare the repository containing the Ops Center Master node backups for use by the 'recovery master' run the command.
      setconfig --recoverymaster <ip_address> --repository <repository_path>

      where <ip_address> is the address of the new master and <repository_path> is the base directory of the repository containing the backup to be recovered. When successful the following response will be issues:

      Repositorylocated at <repository_path> prepared for use by a recovery master

    • Start the protector services. Refer How to start the services.
    • Add Client and Repository nodes to the new recovery master Node, on the UI Nodes screen authorize the client node when it appears. Refer How to authorize a node.
    • Create new node. Refer How to add a node.
    • Select 'Storage' node type "Repository". Refer Node Type Wizard.
    • Specify a temporary name to be used for the repository node.
    • Select the client node as the "Proxy Node"
    • For creation mode select "Import existing repository", giving the path of the repository on the client when asked to "Select existing repository directory".
    • When the node has been created the imported repository will appear in the nodes list and its will appear as 'online' once it has automatically mounted,
  3. Preparing HCP/Cloud Storage for use by the new Master Node

    This section can be ignored if the Ops Center Protector Master Node backups are not stored on a cloud node:

    The new recovery master will be used as the proxy for the storage node. This storage node should be created as follows:

    • Create new node. Refer How to add a node.
    • Select 'Storage' node type and then select the storage type that contains the Ops Center Protector Master back to be recovered. Refer Node Type Wizard.
    • Specify a temporary name to be used for the node.
    • For creation mode select "Import an existing node".
    • For proxy node select the recovery master itself .
    • For the metadata cache directory specify a temporary directory. Be sure that this directory does not already exist.
    • On the configure node screen specify the details of the HCP/Cloud storage.

    When the node has been created the imported storage node will appear in the nodes list and its should appear as 'online' once it has automatically mounted

  4. Recover the Original Master Settings

    • On the restore screen (see How to view available backups) hit the 'Search' button. Unfortunately, it will be difficult to filter the results (other than by specifying the temporary name used for the storage node) because until the master settings have been recovered the node names, policy names, data flow names, etc, used for filtering are not known.
    • Select the most recent master node settings backup from the recovery points listed and hit the 'restore' tool icon (see Repository Snapshot Details (Restore) – Ops Center.
    • Select "Ops Center Protector Master" from the list of applications to restore, and hit 'Next'
    • Select the current (master) node as the destination node and hit the 'Finish' button.
    • The progress of the restore screen may be monitored from the 'Jobs' screen (see How to view and control running jobs) or 'Logs' screen (see How to view logs).
    • When the restore completes it will be shown as 'Completed with warnings'. This is because the process of changing the master node to use the recovered settings cannot be performed while the master node is active. A message will be sent to the Log screen as the job completes detailing  the Windows DOS command/Linux Shell command required to complete the recovery. Take a copy of this command string. For e.g.
      Log Manager - Attachments
      GUID-69B1AEBD-2FDC-4E6B-811A-71F4FEC2B3D3-low.png
    • Stop the protector services. Refer How to stop the services.
    • From a Windows DOS command prompt or Linux shell command run the command noted in the previous step to restore the original master node settings.
    • Start the protector services. Refer How to start the services.

      Where using a repository node the proxy node for that repository will appear on the UI as "offline". To return it to use by the recovered master:

    • Stop the protector services. Refer How to stop the services.
    • To return the repository used for the recovery for use by the 'recovered master' run the command within a Windows DOS/Unix shell window change directory to the Protector 'bin' directory

          

      setconfig --recoveredmaster <ip_address> --repository <repository_path> 
      
      where <ip_address> is the address of the recovered master and <repository_path> is the base directory of the repository. When successful the following response will be issues:

      Repositorylocated at <repository_path> returned for use by the recovered master

    • Start the protector services. Refer How to start the services
    • The master node UI should now be showing the client node and the repository  node as 'online'. 
    NoteAlthough all client nodes will still have copies of their active rules the recovered master does not know what rules are active on each client, and so the UI Monitor screen shows no data flows. All required data flows should now be compiled and activated refer How to activate a data flow.

Revert to Original Master

It is possible, following successful recovery of the master node settings to a new 'Recovered Master', to subsequently revert back to the original master system.

If any changes have been made to the master settings following recovery then a fresh backup of the Ops Center Protector Master Node  settings should be performed and then the same procedure should be followed to restore this backup to the original system as was used to create the recovered master.

If no changes have been made to the master settings following recovery then the following steps may be followed to revert to using the original master node without the need for any new backup or recovery:

  • Stop the protector services (see How to stop the services on the recovered master.
  • Ensure the protector services are not running on the original master .
  • Within a Windows DOS/Unix shell window change directory to the Protector 'bin' directoryon the original master enter the CLI command setconfig --forceprobe
  • Start the cofiohub service on the original master.

 

  • Was this article helpful?