dd860 archiver administration guide - dell emc · dd860 archiver administration guide ... adding...

DD860 Archiver Administration Guide

DD OS 5.1

Backup Recovery Systems DivisionData Domain LLC2421 Mission College Boulevard, Santa Clara, CA 95054866-WE-DDUPE; 408-980-4800

775-0202-0001 Revision AMarch 10, 2011

2

Copyright © 2011 EMC Corporation. All Rights Reserved.

EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without notice.

THE INFORMATION IN THIS PUBLICATION IS PROVIDED "AS IS." EMC CORPORATION MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.

EMC, Data Domain, and Global Compression are registered trademarks or trademarks of EMC Corporation in the United States and/or other countries.

All other trademarks used herein are the property of their respective owners.

Contents

About This Guide. . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Required Additional Data Domain Documents . . . . . . . . . . . 7

Related Data Domain Documents . . . . . . . . . . . . . . . . . . . 7

Access to Documents at Data Domain . . . . . . . . . . . . . . 8

Access Data Domain Documents . . . . . . . . . . . . . . . 8

Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Data Domain Support . . . . . . . . . . . . . . . . . . . . . . . . . . 10

1 Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . .11

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Data Domain Provided Hardware . . . . . . . . . . . . . . . . . . 14

Customer-Provided Infrastructure. . . . . . . . . . . . . . . . . . 14

2 Installation and Configuration . . . . . . . . . . . . . . . .17

Accessing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Configuring the Active Tier . . . . . . . . . . . . . . . . . . . . . . 18

Adding an Archive Tier . . . . . . . . . . . . . . . . . . . . . . . . . 19

Configuring Data Movement. . . . . . . . . . . . . . . . . . . . . . 20

Data Movement Policy . . . . . . . . . . . . . . . . . . . . . . . 20

Common Sizing Errors . . . . . . . . . . . . . . . . . . . . . . . 21

Cleaning and Snapshots . . . . . . . . . . . . . . . . . . . . 21

DD860 Archiver Administration Guide 3

Configuring Replication . . . . . . . . . . . . . . . . . . . . . . . . .23

Collection Replication. . . . . . . . . . . . . . . . . . . . . . . .23

Directory Replication . . . . . . . . . . . . . . . . . . . . . . . .24

MTree Replication . . . . . . . . . . . . . . . . . . . . . . . . . .25

3 Administration . . . . . . . . . . . . . . . . . . . . . . . . . 27

Using the EMC Data Domain Enterprise Manager . . . . . . . . .27

Adding Shelf Capacity Licenses . . . . . . . . . . . . . . . . . .28

Configuring Storage . . . . . . . . . . . . . . . . . . . . . . . . .28

Creating an Archive-Capable File System . . . . . . . . . . .29

Create a New File System: Archive License Applied .29

Data Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . .30

Start or Stop Data Movement on Demand . . . . . . . .30

Modify Data Movement Schedule . . . . . . . . . . . . . .31

Expanding an Archive or Active Tier . . . . . . . . . . . . . .32

Expand Archive Tier . . . . . . . . . . . . . . . . . . . . . .32

Changing Archive Tier Compression . . . . . . . . . . . . . . .32

Modify Archive Tier Compression . . . . . . . . . . . . . .32

About the File System Overview Pane for DD Archiver. . .33

Data Management File System Tab . . . . . . . . . . . . . . .33

Summary Tab . . . . . . . . . . . . . . . . . . . . . . . . . . .33

Archive Units Tab . . . . . . . . . . . . . . . . . . . . . . . .33

Configuration Tab . . . . . . . . . . . . . . . . . . . . . . . .34

Space Usage Tab . . . . . . . . . . . . . . . . . . . . . . . . .34

Consumption Tab . . . . . . . . . . . . . . . . . . . . . . . .34

Daily Written Tab . . . . . . . . . . . . . . . . . . . . . . . .34

Data Management MTree Tab: Data Movement Policy 34

4

Upgrading a DD Archiver. . . . . . . . . . . . . . . . . . . . . . . . 35

Replacing the System Controller . . . . . . . . . . . . . . . . . . 35

Replication Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . 36

DD Archiver Recovery . . . . . . . . . . . . . . . . . . . . . . . . . 37

Overall Recovery Strategy . . . . . . . . . . . . . . . . . . . . . . 38

Reusing an Archive Unit . . . . . . . . . . . . . . . . . . . . . . 38

4 DD Archiver Commands . . . . . . . . . . . . . . . . . . . .39

archive Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

filesys archive commands . . . . . . . . . . . . . . . . . . . . . . . 42

storage Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46


About This Guide

This guide describes how to install, configure, and administer the EMC® Data Domain® Archiver.

Required Additional Data Domain DocumentsThis guide is intended to be used in conjunction with the following documents:

• The Installation and Setup Guide that shipped with your system.

• Data Domain ES20 Hardware Guide

• EMC ES30 Shelf Hardware Guide

Related Data Domain DocumentsThe following Data Domain system documents provide additional information:

• DD OS Release Notes 5.1

• DD OS 5.1 Initial Configuration Guide

• DD OS 5.1 Administration Guide

• DD OS 5.1 Command Reference Guide

• Data Domain Hardware Guide (specific to your system)


Access to Documents at Data DomainThe Documentation page at https://my.datadomain.com/documentation provides access to three categories of documents that are related to use of Data Domain products:

• End-user documents, under Product Documentation.

• Guides for how to integrate Data Domain systems with backup applications, arrays, and other applications, under Integration Documentation.

• Matrices that show which components are compatible with each other, under Compatibility Matrices:

• Data Domain hardware product numbers

• Data Domain operating system (DD OS) versions

• Backup applications and other application versions

• Application server and client operating system versions

• Hardware driver versions

Access Data Domain Documents

1. Log into the support portal at: https://my.datadomain.com/documentation.

2. To view end user documents, click Product Documentation and then perform the following steps:

a. Select the Data Domain model from the Platform list and click View.

b. On the row for the correct Data Domain operating system (DD OS) version, click View under Documentation.

c. Click the desired title.

8 About This Guide

https://mydatadomain.com/documentation

3. To view integration-related documents, perform the following steps:

a. Click Integration Documentation.

b. Select a vendor from the Vendor menu.

c. Select the desired title from the list and click View.

4. To view compatibility matrices, perform the following steps:

a. Click Compatibility Matrices.

b. Select the desired title from product menu and click View.

ConventionsThe following table describes the typographical conventions used in this guide.

Typeface or Symbol

Usage Examples

Monospace Commands, command options, and parameters and computer output.

Use the config command to manage the Data Domain system configuration settings.

Monospace bold

Commands the user types at the command prompt (#).

Enter:# config setup

Monospace italic bold

Command variables the user types at the command prompt (#).

# log view file_name

Italic Book titles, and variables. Refer to the DD OS Command Reference Guide for complete descriptions of DD OS commands.

Pipe (|) and curly braces ({})

Choose (pipe) between a required argument (curly braces) in the CLI.

{arg1 | arg2}

Brackets ([]) and ellipses (...)

One or more (list with commas and elipses) optional (bracket) arguments in the CLI.

[arg1, arg2, ...]


AudienceThis guide is for administrators and service providers who install, configure, and administer DD860 Archiver.

Data Domain SupportTo resolve issues with Data Domain products, contact your contracted support provider or visit us online athttps://my.datadomain.com.

10 About This Guide

1 Getting Started

This chapter covers the following topics:

• Introduction on page 11

• Licenses on page 13

• Data Domain Provided Hardware on page 14

• Customer-Provided Infrastructure on page 14

IntroductionThe EMC Data Domain Archiver (DD Archiver) addresses the need to cost-effectively retain and access deduplicated data for longer periods. Instead of storing long-term backup data on tape, it can be stored locally on the Archiver. The result is lower operational cost as the deduplication engine can be shared across more shelves and managed as a unit.

Two-tiered File System

DD Archiver provides an internal two-tiered file system—active and archive. (The file system appears to users as a single entity.) Incoming data to the DD Archiver is first placed in the active tier of the file system. The data (in the form of complete files) is later moved to the archive tier of the file system as specified by data movement policies. For example, the active tier might retain weekly full and daily incremental backups for 90 days while the archive tier might retain the monthly fulls for seven years.

Transparency of Operation

DD Archiver supports existing applications via standard protocols with WORM capabilities (Retention Lock) and automatic transparent data movement from the active tier to the archive tier. Applications operate on the DD Archiver as they do on a standard


Data Domain system. All of the data in the two tiers is accessible, although there might be a slight delay on initial access to data in the archive tier. The namespace of the system is global, and not affected by data movement. No partitioning of the file system is necessary to take advantage of the two-tiered file system.

Data Movement Policies

For DD OS 5.1, the data movement policies are based on when a file was last modified. This date and time can be set differently for subsets of data. Files that may be updated for a period of time need a different policy than ones that never change.

Deduplication

The archive tier is comprised of one or more archive units, each of which may consist of one or more entire shelf’s worth of storage. For fault isolation purposes, deduplication occurs within each archive unit. There is no cross-deduplication among archive units.

Storage

The concept of tiering extends to the storage level. The active tier of the file system draws storage from the active tier of storage. The archive tier of the file system uses storage from the archive tier of storage. For both tiers, DD OS 5.1 supports the ES20 shelves with 7200 RPM disks and the 1 and 2 TB disk drive models of ES30 shelves with 7200 RPM disks.

Note: A DD Archiver system can use both ES20 and ES30 models of shelves. Different models cannot be mixed in the same shelf set, and the shelf sets must be balanced according to the configuration rules specified in the EMC ES30 Shelf Hardware Guide.

The DD Archiver enables you to attach significantly more storage to the same controller, up to a maximum of 24 shelves with 570 TB of total capacity. The active tier must include at least one shelf worth of storage. As one archive unit fills up, simply add an additional archive unit. DD Archiver allows incremental capacity growth for the long-term retention of data.

Data Protection

Data is protected with built-in fault isolation features, disaster recovery capability, and Data Invulnerability Architecture (DIA).

12 Getting Started

DIA checks files when they are moved from the active to the archive tier. After data is copied into the archive tier, the container and file system structures are read back and verified. The location of the file is updated, and the space on the active tier is reclaimed only after the file is verified to have been correctly written to the archive tier.

When an archive unit is filled up, namespace information and system files are copied into it so that the data in the archive unit may be recovered even when other parts of the system are lost. An archive unit is completely filled up and sealed before another one is used.

Note: Encryption, sanitization, and data access via VTL are not supported for this release.

LicensesEMC Data Domain Archiver consists of a DD860 controller with the DD Archiver configuration, the DD Archiver license, and a shelf capacity license for each storage shelf. The DD Archiver license is required to enable the DD Archiver feature. The DD Archiver feature can be enabled only on a system without a file system. If the DD Archiver feature is not enabled, the system behaves as a standard DD860, which has a maximum usable storage capacity of 140 TB.

The appropriate shelf capacity license is required for any new shelf that you add. The license is specific to either an active or archive tier shelf. An Expanded-Storage license is required to expand the Active Tier Storage above 64 TB of raw capacity; for example, more than two ES20-32 TB shelves. You cannot use additional storage without first applying the appropriate licenses.


Data Domain Provided HardwareThe DD Archiver system consists of a DD860 controller with the DD Archiver configuration, which includes:

72 GB of RAM

1 - NVRAM card (1 GB)

3 - quad-port SAS cards

2 - 1 GbE ports on the motherboard

0 to 2 - 1/10 GbE cards for external connectivity

1 to 24 - ES20 or ES30 shelves (1 TB or 2 TB disks)

Notes:

After the DD Archiver feature is enabled, the maximum usable storage capacity of an active tier is 142 TB. The maximum usable storage capacity of an archive tier is 544 TB. The active and archive tiers have a total usable storage capacity of 568 TB.

Upgrade of the DD860 controller to the DD Archiver configuration is not supported in the field.

Customer-Provided Infrastructure• Specifications and Site Requirements

See the DD860 Archiver Installation and Setup Guide.

• Rack Space and Interconnect Cabling

See the DD860 Archiver Installation and Setup Guide.

• Racking and Cabling

Rack the system with future expansion (up to 24 ES20 or ES30 shelves) in mind. All of the shelves are attached to the single Data Domain system. It is recommended that the initial system configuration be racked as shown in the Data Domain ES20 Hardware Guide for a system consisting of only ES20 shelves, or the EMC ES30 Shelf Hardware Guide, for a system with ES30 shelves or combined ES20 and ES30 shelves.

14 Getting Started

The recommended racking procedure is as follows:

• Install the base cabinet, including the DD860 controller, in the rack space at RU13 and RU14.

• Install the initial shelves at the bottom of the cabinet.

• Add shelves from the bottom up until the rack is full.

• Completely fill up a set before creating another set.


16 Getting Started

2 Installation and Configuration

The steps to install and configure DD Archiver are similar to those for a regular Data Domain 860 controller. Follow the instructions given in the Installation and Setup Guide for the EMC Data Domain DD860 Archiver Storage System.

After the DD860 setup is complete, implement the DD Archiver-specific procedures given in this chapter.

The chapter covers the following topics:

• Accessing Data on page 17

• Configuring the Active Tier on page 18

• Adding an Archive Tier on page 19

• Configuring Data Movement on page 20

• Configuring Replication on page 23

Accessing DataDD Archiver supports the protocols NFS, CIFS, and DD Boost. For a list of applications supported with DD Boost on the DD Archiver, see the DD Boost Compatibility List on the Data Domain support site.

Data first lands in the active tier. Files are moved in their entirety into archive units in the archive tier as specified by data movement policies. All files appear in the same namespace. There is no need to partition data, and you can continue to expand the file system as desired.

All data is visible to users, and all file system metadata is present in the active tier. All data is accessible via CIFS, NFS and DD Boost.


The trade-off in moving data to archive tiers is larger capacity versus slightly slower access time if the unit to be accessed is not currently ready for access.

Configuring the Active TierThe DD Archiver cannot start writing data without a configured active tier. The active tier is analogous to a standard Data Domain system and should be sized similarly. The maximum amount of storage that can be added to the active tier is 142 TB.

Note: For more information about archiver-related commands, see Chapter 4, DD Archiver Commands.

Install licenses and enable DD ArchiverThe DD Archiver license is required. Enable the DD Archiver feature before creating the file system. You cannot convert between a regular file system and a DD Archiver file system.

1. Install the DD Archiver license.

# license add archiver-license-key

2. Enable the DD Archiver feature.

# archive enable

Add storageNotes:

Each storage shelf requires a shelf capacity license. This license is specific for either active or archive tier usage of the shelf.

An Expanded-Storage license is required to expand the active tier storage above 48 TB.

1. To add shelf capacity licenses, enter the license add command followed by one or more license keys, separated with a space. For example:

# license add Capacity-Active-license-keys Capacity-Archive-license-keys


2. Issue this command once for each enclosure that is to be added to the active tier:

# storage add tier active enclosure enclosure-id

Create and enable the file system1. Create the file system.

# filesys create

2. Enable the file system.

# filesys enable

3. Verify that file system is up and running.

# filesys status

When the file system is up and running, the DD Archiver is ready to start servicing read and write requests.

Adding an Archive TierEach archive unit also is a fault-isolation domain. If a unit is damaged, only the files on that unit are affected. The trade-off is while a larger unit size enhances deduplication and performance, it increases the amount of data compromised should a unit fail.

Add an archive unit1. Issue this command for all enclosures to be added:

# storage add tier archive enclosure enclosure-id

2. Create an archive unit and add it to the file system. Issue this command for each archive unit to be added.

# filesys archive unit add

3. Verify that the archive unit is created and added to the file system.

# filesys archive unit list all


Configuring Data MovementA file is moved from the active to the archive tier based on the date it was last modified. For data integrity, the entire file is moved at this time.

You can specify the data movement age threshold, such as nine weeks, after which data that has not been modified is to be moved from the active to the archive tier.

You can specify different age thresholds for each of your defined MTrees. An MTree is a subtree within the namespace that is a logical set of data for management purposes. For example, you might place financial data, emails, and engineering data in separate MTrees.

Data Movement PolicySet the data movement age threshold to distinguish between data that is changing and data that is static and should be archived.

For example, if the retention period of daily-incremental backups is eight weeks and the retention period of weekly fulls is three years, set the age threshold to nine weeks. If you set it lower, you would be moving daily incremental data that is soon to be deleted.

Segregate data with different degrees of change into different MTrees and set the data movement age threshold to move the data soon after the data stabilizes. For example, create MTree A for the daily-incremental backups and create MTree B for the weekly fulls. Set the age threshold for MTree A such that the data in MTree A is never moved and the age threshold for MTree B to one week.

Note: Ensure that the active tier is large enough to hold the data that is changing.


Common Sizing ErrorsAvoid these common sizing errors:

• Setting a data movement policy that is overly aggressive in which data is moved too soon.

• Setting a data movement policy that is too conservative. Once the active tier fills up, no more data can be written to the system.

• Defining an undersized active unit, such that the active tier fills up prematurely.

• Creating an overly aggressive movement policy to compensate for an undersized active tier. Space is not regularly reclaimed in the archive tier so moving files that are to be deleted or updated soon into the archive tier results in wasted space.

Cleaning and Snapshots

Cleaning is performed on the active tier either as scheduled, or by default immediately after files have been moved from the active to the archive tier.

Files in snapshots are not cleaned, even after they have been moved to the archive tier. The space cannot be reclaimed until the snapshots have been deleted. For a DD Archiver system, set the retention for snapshots to less than two weeks.


Configure data movement1. Set the age threshold:

# archive data-movement policy set age-threshold duration mtrees mtree-list

The duration is expressed in number of days. A default age threshold of 14 days applies to MTrees for which the age threshold has not been explicitly set.

# archive data-movement policy set default-age-threshold period

2. Verify the age threshold setting.

# archive data-movement policy show [mtrees mtree-list | all ]

3. Specify the migration schedule.

The movement of files from the active tier to the archive tier is performed periodically based on a schedule that can be specified. It is recommended that file system cleaning be performed after the movement, which is the default behavior.

# archive data-movement schedule set {days days time time [no-clean]} |never

Acceptable schedule values include:

days sun time 00:00

days mon,tue time 00:00

days 2 time 10:00

days 2,15 time 10:00

days last time 10:00 - last day of the month

4. Verify the migration schedule.

# archive data-movement schedule show


Configuring Replication To protect the DD Archiver in case of a disaster, it is recommended that you replicate the data to a second DD Archiver at a remote site using collection replication.

Collection ReplicationCollection replication takes place between corresponding active tier and archive units. If the active tier or an archive unit at the source fails, the data can be copied from the corresponding unit at the remote site onto a new unit, which is shipped to your site as a replacement unit.

Prerequisites for setting up collection replication:

• Both the source and destination systems must be configured as DD Archivers.

• The destination system must have the same number (or more) corresponding archive units as at the source.

• Each unit at the destination must be equivalent or larger in capacity than its corresponding source unit.

• The file system must not be enabled on the destination until the archive units have been added to it, and replication has been configured as described below.

Configure collection replication1. Install the replication license on both the source and

destination.

# license add replication-license-key

2. Configure replication on both the source and destination. The file system must not have been enabled on the destination before this step.

# replication add source col://hostA destination col://hostB

3. Enable the file system on the destination system. On the destination system, enter:


# filesys enable

4. Start the replication on the source.

# replication initialize col://hostB

5. Check the replication status.

# replication status

Directory ReplicationTo copy data into DD Archiver using directory replication, you need a DD OS 5.0 replication source. Replicating from releases prior to DD OS 5.0 requires that the data first be imported into an intermediate Data Domain system running DD OS 5.0. For example, replication from a system that runs DD OS 4.9 is made into a non-DD Archiver Data Domain system that runs DD OS 5.0. Then replication is made from the DD OS 5.0 Data Domain system into the DD Archiver.

Copy data into DD Archiver using directory replication

1. Set up directory replication into the /backup directory on DD Archiver.

2. On the DD Archiver, perform a fast copy from the /backup directory into the target MTree.

3. Manage each MTree with per-MTree data movement policies.

Notes:

Ingest the data in age-order (oldest data first) to maximize deduplication and to improve performance.

If using DD Boost 2.3 or later, you can specify how multiple copies are to be made and managed within the backup application.


MTree ReplicationMTree replication takes place between corresponding active tier archive units. If the active tier or an archive unit at the source fails, the data can be copied from the corresponding unit at the remote site onto a new unit, which is shipped to your site as a replacement unit.

Copy data into DD Archiver using MTree Replication

1. For each MTree to be copied, set up MTree replication into a corresponding MTree on the DD Archiver.

2. On the DD Archiver, perform a fast copy from each corresponding MTree into another MTree for long-term retention.

3. Manage each MTree for long-term retention with per-MTree data movement policies.


3 Administration

This chapter covers the following topics:

• Using the EMC Data Domain Enterprise Manager on page 27

• Upgrading a DD Archiver on page 35

• Replacing the System Controller on page 35

• Replication Recovery on page 36

• DD Archiver Recovery on page 37

• Overall Recovery Strategy on page 38

Using the EMC Data Domain Enterprise Manager

You can use the Enterprise Manager instead of administrative archive commands, except for the functionality provided by the archive report command and the filesys archive command’s delete and expand archive unit options. Access these commands using the command-line interface.

The following topics are covered:

• Adding Shelf Capacity Licenses on page 28

• Configuring Storage on page 28

• Creating an Archive-Capable File System on page 29

• Data Movement on page 30

• Changing Archive Tier Compression on page 32

• About the File System Overview Pane for DD Archiver on page 33

• Data Management File System Tab on page 33


Adding Shelf Capacity Licenses1. Within the Enterprise Manager, open the System Settings >

Licenses tab.

2. Click Add Licenses.

3. Enter one or more licenses, one per line.

Note: If there are any errors, a summary of the added licenses and those not added because of error are listed. Click Fix Errors and reenter the licenses keys. Click Add.

The licenses for the Data Domain system are displayed in two groups:

• Feature Licenses, required for features such as DDBoost

• Shelf Capacity Licenses, which displays shelf capacity in TiB, shelf model, such as ES30, and the shelf’s storage tier (active or archive).

To delete a license, select the license in the Licenses list and click Delete Selected Licenses. If prompted to confirm, read the warning and click OK to proceed.

Configuring StorageNote: Additional storage requires the appropriate license or licenses, and the Data Domain system must have enough installed memory to support it. Error messages display if more licenses or memory is needed.

1. Within the Enterprise Manager, open the Hardware > Storage tab.

2. In the Overview tab, click Configure Storage.

3. In the Configure Storage tab, select the storage to be added from the Available Storage list.

4. Select the appropriate Tier Configuration (Archive or Active) from the menu.

Note: The two bars show the portion of licensed capacity used/remaining for each shelf model (ES20 and ES30).

28 Administration

5. Select the check box for the Shelf to be added.

6. Click the Add to Tier button.

7. Click OK to add the storage.

Note: To remove an added shelf, select it in the Tier Configuration list and click Remove from Configuration. Click OK.

Creating an Archive-Capable File System

Create a New File System: Archive License Applied

Note: Add the archive license to the DD Archiver before using this procedure.

1. Within the Enterprise Manager, open the Data Management > File System tab.

2. Select Create file system from the More Tasks menu.

Note: Create file system is not available as an option in the More Tasks menu if a file system is up and running.

3. Storage is classified in tiers—either active tier or archive tier. Select the type of file system to be created: either an archive capable file system (the default), or a non-archive file system (to implement DD Archiver at a later date).

Note: After one of these two file systems has been created, it cannot be converted to another type without destroying the current file system.

4. Click Next.

5. Storage must be configured before the file system can be created. Click Configure in the File System Create dialog box.

6. When the Configure Storage Status dialog box shows the disk enclosures addition has completed, click Close.

7. The storage in the active tier is used to create the active file system tier. The storage in the archive tier is used to create archive units. Using the File System Create dialog box, you can choose to create an active tier and the first archive unit. Click Next to confirm.


Note: You want the file system to be enabled (the default) after the file system has been created.

8. Select the size of the first archive unit from the menu.

9. Click Next.

A Summary page shows the size of the active and archive tiers in the new file system. You have the option of returning to the previous step to change the allocation size, or of cancelling the procedure.

10. Click Finish to create the file system.

The progress of each step is shown, and a progress bar monitors overall status.

Data MovementThe File System page for a DD Archiver system shows the status of data as it is moved from the active to the archive tier. The target unit in the archive tier is the recipient of the data. The status includes when the data movement completed, the number of files copied, and the amount of data copied in GiB.

Clicking the Data Movement Status Start button starts the data movement based on your defined data movement policy.

Start or Stop Data Movement on Demand

1. Go to Data Management > File System.

2. Click the Start button to the right of Data Movement Status.

3. The Start Data Movement dialog box warns that data is to be moved from the active to the archive tier as defined in your data movement policy, followed by a file system clean. Click Finish to start the data movement.

Note: If cleaning is already in progress, starting the data movement schedules the data movement to run after the clean completes.

The data movement status is shown in the File System tab. The Start button is replaced by a Stop button.

30 Administration

4. Clicking Stop stops the data movement. Click OK in the Stop Data Movement dialog box to confirm.

Modify Data Movement Schedule

1. Go to Data Management > File System and select the Configuration tab.

2. Click the Data Movement Policy’s Edit button.

a. In the Data Movement Policy dialog box, select the threshold number of days. When the data movement starts, all of the files that have not been accessed for the specified threshold number of days are moved from the active tier to the archive tier.

b. Set a schedule for when the data movement should begin; for example, weekly, daily, specific days, and a time in hours and minutes.

c. A file system clean is recommended after the data movement. De-select this option if you do not want the clean to occur.


Expanding an Archive or Active TierWhen a file system exists, you can expand an archive or active tier.

Expand Archive Tier

1. To expand an archive or active tier, select the Data Domain system and go to Data Management > File System.

2. From the More Tasks menu, select Expand File System.

3. In the Expand File System Capacity dialog box, you have the option of expanding either the active or archive tier. Make your selection and click Next.

• To expand an archive file system, you need to create a new archive file system unit. In the Expand File System Capacity dialog box, select the size of the new archive unit. Click Configure.

• To expand an active file system, click Configure.

4. After the configuration completes, click Finish.

Changing Archive Tier CompressionTo display archive tier compression statistics, issue the filesys show compression command.

Modify Archive Tier Compression

1. Go to Data Management > File System and select the Configuration tab.

2. Click the Options Edit button.

3. Select one of the compression options from the Archive Tier Local Compression menu, and click OK.

The default is gz, which is a zip-style compression that uses the least amount of space for data storage (10% to 20% less than lz on average; however, some data sets achieve much higher compression).

See the DD OS Administration Guide for a description of other compression options.

32 Administration

About the File System Overview Pane for DD Archiver

The File System overview pane for a DD Archiver system displays the file system Data Movement Status, which is the time that the last data movement finished. Pressing the Start button enables you to start the data movement. When the data movement is running, the Start button changes to a Stop button. Pressing the Stop button stops the data movement.

Data Management File System Tab

Summary Tab

The File System Summary tab displays information about disk space usage and compression for both the active and archive tiers.

Disk Space Usage: This pane shows the total size, amount of space used, and amount of available space and combined totals for active and archive tiers. The amount of cleanable space is shown for a active tier.

Compression: Shows the pre-compression and post-compression values that are currently used and those that have been written in the last 24 hours. Also shows the global, local, and total compression (reduction percentage) factors.

Archive Units Tab

The Archive Units tab on the File System page lists each archive unit. It shows the unit’s state (new, sealed, or target), its status (disabled, ready, or stand-by), and its size. If the unit has been sealed, that is, no more data can be added, then the date that it was sealed is given.

Click the diamond symbol to the right of a column heading to sort the order of the values in reverse.


Configuration Tab

Modify Basic Settings: Compression

Clicking the Option’s Edit button allows you to modify this setting. The compression options are none, lz, gz (the default), and gzfast.

Space Usage Tab

You can select one of the three chart types to view space usage over time in MiB for the entire file system, or for an archive or active tier. Select one of the duration values in the upper-right of the tab. The data is presented (color-coded) as pre-compression written (blue), post-compression used (red), and the compression factor (black).

Consumption Tab

The consumption graph displays the amount of post-compression storage used and the compression ratio over time, which enables you to view consumption trends. An option allows you to display the post-compression storage against total system capacity.

Daily Written Tab

Shows the amount of data written per day for a selected duration (7, 30, 60, or 120 days). The data is presented (color-coded) in both graph and table format as pre-compression written (blue), post-compression used (red), and the compression factor (black).

Data Management MTree Tab: Data Movement Policy

The Data Movement Policy shows the current threshold for the age of a file before it is archived. Fourteen days is the default. To change the value, click Edit, enter the new value, and click OK.

34 Administration

Upgrading a DD ArchiverThe DD Archiver follows the same upgrade policy as a standard Data Domain controller. Upgrading from up to two major prior releases is supported. The upgrade process is initiated by the following command where file specifies an rpm in the /ddr/var/releases directory:

# system upgrade file

If the active tier is not available, the upgrade process upgrades the system chassis and places it into the state where it is ready to create or accept a file system. Otherwise, the process proceeds to upgrade the active tier and all available archive units, and puts the system into the state that the previous upgrade has not been verified to be complete. This state is cleared by the file system after the file system is enabled and has verified that all archive units belonging to the file system have been upgraded. A subsequent upgrade is not permitted until this state is cleared.

If an archive unit becomes available after the upgrade process has finished, the unit is automatically upgraded when it is plugged into the system, or at the next system start.

Replacing the System ControllerThe following conditions must be met to replace one DD Archiver system controller with another:

• The new system controller must not already own an active tier or archive unit.

• The active tier and all archive units to be owned by the new system controller must be at the same version. Any active tier or archive unit that is at a different version must not be connected to the new system controller until the replacement process is completed.

A system controller replacement affects DD Archiver as follows:

• If the new system controller is at a more recent version than the active tier and archive units, the active tier and archive units are upgraded to that system controller’s version. Otherwise,


the new system controller is upgraded to the version in the active tier and archive units.

• The active tier and archive units that are connected to the new system controller become owned by the new system controller. All visible active tier and archive units are attached to the new system controller.

• If there is an active tier, the registry in the active tier is installed in the new system controller. Otherwise, the registry in the archive unit with the most recently updated registry is installed in the new system controller.

Start the chassis replacement procedureEnter:

# system headswap

Replication RecoveryYou can use any available collection replica to recover from a disaster affecting the source DD Archiver.

Notes:

If you do not have collection replica, see DD Archiver Recovery on page 37.

If you need to recover only a portion of the system, such as one archive unit, from the collection replica, contact Data Domain Support.

The new source must be configured as a DD Archiver with the same number (or more) archive units as are located at the destination. The file system must not be enabled on the new source until the archive units have been added to it and replication recovery has been initiated.

1. Install the replication license on the new source.

# license add replication-license-key

2. Reset the authentication key on the destination.

# replication reauth col://hostB

36 Administration

3. Reconfigure replication on both the new source and destination.

# replication add source col://hostC destination col://hostB

4. Initiate recovery on the new source. The file system must not have been enabled on the new source before this step.

# replication recover col://hostB

5. Check the replication status.

# replication status

DD Archiver RecoveryIf the active tier and a subset of the archive units are lost and there is no replica available, Data Domain Support may be able to reconstitute any remaining sealed archive units into a new DD Archiver.

The DD Archiver is designed to remain available to service read and write requests when one or more archive units are lost. The file system may not detect that an archive unit is lost until the file system restarts or tries to access data stored in the archive unit. The latter circumstance may trigger a file system restart. After the file system has detected that an archive unit is lost, it returns an error in response to requests for data stored in that unit.

If the lost data cannot be recovered from a replica, Data Domain Support might be able to clean up the system by deleting the lost archive unit and any files that reside fully or partially in it.


Overall Recovery StrategyThe DD Archiver is equipped with tools to address failures in different parts of the system. In a failure situation, perform recovery actions in the order listed below.

Recover from DD Archiver failures1. Restore connection between the system controller and storage.

If the system controller is lost, replace it with a new system controller.

2. If there is loss of data and a replica is available, try to recover the data from the replica.

If a replica is not available, limit any loss of data by leveraging the fault isolation features of the DD Archiver through Data Domain Support.

Reusing an Archive UnitIf all of the files on an archive unit are no longer needed, deleting them makes the unit available for reuse. You can generate a file location report to make sure that the archive unit is indeed empty, delete the archive unit, then add it as a new archive unit.

38 Administration

4 DD Archiver Commands

This chapter covers the following DD Archiver-related commands.

• archive Command on page 39

• filesys archive commands on page 42

• storage Command on page 46

archive CommandUnless otherwise noted, commands require administrator privileges.

archive data-movement policy reset age-threshold mtrees mtree-list

Reset the age threshold for specified MTrees to the default value of 14 days. Files modified only in the past (beyond the age threshold) are moved to the archive tier in the next data movement.

archive data-movement policy reset default age-threshold

Reset the default age threshold to the default value of 14 days. The default age threshold applies to MTrees for which the age threshold has not been set.

Sample output:

The default-age-threshold value is set to "14" days.

archive data-movement policy set default-age-threshold days mtrees mtree-list

Set the age threshold for the specified MTrees.

archive data-movement policy set default-age-threshold days

Set the default age threshold.


archive data-movement policy show [mtrees mtree-list]

View the data-movement policy for the specified MTrees.

archive data-movement schedule reset

Reset the data-movement schedule.

Sample output:

The Archive data-movement schedule has been reset.

Archive data movement is scheduled to run on day(s) "Tue" at "06:00"

archive data-movement schedule set {days days time time[no-clean]} | never

Set the schedule for data movement. Acceptable values include:

days sun time 00:00

days mon,tue time 00:00

days 2 time 10:00

days 2,15 time 10:00

days last time 10:00 - last day of the month

Sample output:

The Archive data-movement schedule has been set.

Archive data-movement is scheduled to run on day(s) "tue" at "06:00" hrs

Unless you specify no-clean, the file system is cleaned after data movement is completed.

archive data-movement schedule show

Display the data-movement schedule. This command may be run without administrator privileges.

archive data-movement schedule status

Display the data-movement status. This command may be run without administrator privileges.


archive data-movement start

Start data movement from active tier to archive tier. All files that satisfy the age-threshold value are moved to the archive tier.

Sample output:

Waiting for data movement to start...

Data movement has started. Run "filesys data-movement watch" to monitor the progress.

archive data-movement stop

Stop data movement to the archive tier.

Sample output:

Waiting for data movement to stop...

Data movement has stopped.

archive data-movement watch

Display data-movement progress. This command may be run without administrator privileges.

archive disable

Disable archive feature.

archive enable

Enable archive feature.

archive option reset local-compression type

Reset the local compression algorithm to the default of gz for subsequent data movement to the archive tier.

Sample output:

The local compression for archive tier is reset. The file system must be restarted to effect this change.

archive option set local-compression-type (none|lz|gzfast|gz}

Set the local compression algorithm for subsequent data movement to the archive tier.


Sample output:

The local compression will be set to gz. The file system must be restarted to effect this change.

archive option show local-compression-type

Display the local compression algorithm for the archive tier. This command may be run without administrator privileges.

archive report generate file-location

Create a report showing the name of each file in the system and the location; for example, archive unit ar1. The report is saved in the specified file location.

An asterisk (*) is appended to the archive unit name if the file spans the active tier and the archive unit.

Press Ctrl-C to stop generating the report.

archive show config

Display archive configuration. This command may be run without administrator privileges.

filesys archive commands

filesys archive unit add

Create an archive unit from the archive tier of the file system, change the state of disks or LUNs from “available” to “in use,” and add the new archive unit to the file system. Also view the list of archive unit sizes available in the archive tier.

In the following example, the command lists four archive unit sizes, the administrator selects the second size listed, and the command creates archive-unit-16:

# filesys archive unit add


The archive unit can be created with the following estimated sizes:

10.90TiB

21.80TiB

32.70TiB

43.60TiB

Enter the size of the archive unit: 21.80TiB

Creating Archive unit.

########################################### [5%]

########################################### [55%]

Archive unit "archive-unit-16" is created.

filesys archive unit del

Delete the specified archive unit and change the state of disks or LUNs to “available.” An archive unit can be deleted from the tier only when the Data Domain system file system is disabled. Disabling the file system stops all Data Domain system operations, including a filesys clean operation.

Notes:

This command destroys all data in the specified archive unit.

Any file in the specified archive unit must be deleted to remove it from the namespace.

In the following example, the user is prompted to disable the file system and retry the command. Before deleting archive-unit-6, the user must acknowledge a warning message and enter the sysadmin password.

# filesys archive unit del archive-unit-6

The archive unit can be deleted only when the filesystem is disabled.

Run "filesys disable" and retry the operation.

# filesys disable

Please wait.................


The filesystem is now disabled.

# filesys archive unit del archive-unit-6

The 'filesys archive unit del' command irrevocably destroys all data in the archive unit "archive-unit-6".

Are you sure? (yes|no|?) [no]: yes

ok, proceeding.

Please enter sysadmin password to confirm 'filesys archive unit del':

Destroying archive unit "archive-unit-6"

########################################### [100%]

Archive unit "archive-unit-6" is destroyed.


Name State Status Archive Archive Size DiskgroupsStart Time End Time (GiB)

-------------- ------ ------ ------------ ------------ ----- ----------archive-unit-6 Sealed Ready Fri ... 2010 Mon ... 2010 33055 dg4,dg3-------------- ------ ------ ------------ ------------ ----- ----------

# filesys enable

filesys archive unit expand

Expand the size of the specified archive unit.


filesys archive unit list [ | all]

List all archive units or a specific archive unit.

Output includes the following information for each archive unit:

• Name

• State

• New—Available as a target archive unit.

• Target—Current destination of data movement.

• Sealed—Unavailable as a target archive unit because it has reached maximum capacity of archived data.

• Status : Ready or Offline

• Archive Start Time

• Archive End Time

• Size (in GiB)

• Diskgroups

The following example displays all archive units in a DD Archiver system file system:


Name State Status Archive Size DiskgroupStart Time End Time (GiB)

-------------- ------ ------ -------------------------------- ------ ------archive-unit-1 Sealed Ready 2006-12-09 12:50 2008-12-09 12:50 10000 dg1archive-unit-2 Target Ready 2008-12-09 12:50 N/A 10000 dg2archive-unit-3 New Ready N/A N/A 10000 dg3


storage Commandstorage add [tier active] {enclosure enclosure-id | devlun-id | disk enclosure-id.disk-id [spindle-group 1—16]}

storage add tier {active | archive} {enclosure enclosure-id | devlun-id | disk enclosure-id.disk-id [spindle-group 1—16]}

Add the specified storage devices to the storage tier. This command is available to administrative users only. The devices you add to a tier can be all disks in an enclosure, a single disk, or a LUN in a gateway system. This command takes unknown disks or LUNs and makes them available in the specified tier.

For example:

#storage add enclosure 2 tier archive

Checking storage requirements...done

Adding enclosure 2 to the archive tier...done

Updating system information...done

Enclosure 2 successfully added to the archive tier

Note: This command does not apply to the disks in a dataless head (DLH) unit.

This command supports the following options:

tier {active | archive} In a single-node Data Domain system, the specified storage devices are added to the active tier by default. In a DD Archiver system only, you must specify whether the storage devices are added to the active tier or the archive tier.

enclosure (Optional) Take all the disks in the specified enclosure to the tier, provided all the disks are in the unknown state, and make them available in the tier. This is the preferred method of adding storage to a Data Domain system.


storage remove {enclosure enclosure-id | devlun-id | disk enclosure-id.disk-id}

Remove the specified storage devices from a storage tier or dataless head (DLH) unit. This command is available to administrative users only.

You can remove storage device—all disks in an enclosure, a single disk, or a single LUN in a gateway system— from a tier. You can also remove a single disk from a DLH unit. This command removes disks or LUNs from a storage tier and changes the device state to unknown.


disk (Optional) Take the specified unknown disk and make it available in the tier. This option is allowed only after the enclosure has been added via the storage add enclosure command.

dev (Optional) Take the specified unknown LUN in a gateway system and make it available in the tier.

spindle-group (Optional) Specify the spindle group to which the LUN is assigned. By default, the LUN is assigned to spindle group 1 is added.

enclosure (Optional) Remove all disks in the specified enclosure from the tier to which they belong. The enclosure must not contain in use disks.

disk (Optional) Add the specified spare or in use disk from the tier or DLH unit to which it belongs. Note: This command will not remove an in use disk from a tier if doing so would reduce the number of disks of a RAID group beyond the minimum for the RAID scheme in the Data Domain system.


storage show {all | summary | tier active}

storage show {all | summary | tier {active | archive}}

Display a storage summary and tier-specific storage details for the file system.

The storage summary shows the number of disks in each tier by disk state.

For each tier, the storage details show the count and identifiers of each disk in the tier by disk group. The details include the total number of disk groups working normally and the total number of disk groups that are degraded. If disk groups in the tier are queued for reconstruction or undergoing reconstruction, details of the reconstruction progress are also displayed.

By default, this command displays the storage summary and also storage details for each tier in the file system. To limit the command output, include command options.


dev (Optional) Remove the specified available LUN in a gateway system from the tier to which it belongs. Note: This command will not remove a spare or in use LUN from a storage tier.

all Display the storage summary and also storage details for each tier.

summary Display only the storage summary.

tier active Display only the storage details for the active tier.

tier archive In a DD Archiver system only, display only the storage details for the archive tier.


For following example command shows output for the storage show all command.

# storage show all

Active tier details:

Disk Disks Count Disk Additional

Group Size Information

------- -------- ----- ------- -----------

dg1 5.1-5.14 14 1.8 TiB

(spare) 5.15 1 1.8 TiB

------- -------- ----- ------- -----------

Archive tier details:



----------- -------- ----- --------- -----------

(available) 2.1-2.16 16 465.7 GiB

(available) 3.1-3.16 16 931.5 GiB

----------- -------- ----- --------- -----------

Storage addable disks:

Disk Disks Count Disk Enclosure Shelf Capacity Additional

Type Size Model License Needed Information

--------- -------- ----- --------- --------- -------------- -----------

(unknown) 4.1-4.15 15 465.7 GiB ES20 5.4 TiB

(unknown) 6.1-6.16 16 465.7 GiB ES20 N/A

--------- -------- ----- --------- --------- -------------- -----------


Shelf Capacity License:

License Model Total Used Remaining

---------------- ----- -------- -------- ---------

CAPACITY-ACTIVE ES20 16.3 TiB 0.0 TiB 16.3 TiB

CAPACITY-ACTIVE ES30 21.8 TiB 21.8 TiB 0.0 TiB

CAPACITY-ARCHIVE ES20 16.3 TiB 5.4 TiB 10.9 TiB


---------------- ----- -------- -------- ---------

The following example command displays the number of disks in each tier by disk state:

# storage show summary

Disk states Active tier Archive tier Other--------------------- ----------- ------------ -----in-use 27 28 -spare 3 4 -spare(reconstruction) 1 - -available 16 - -unknown - - 15fail - - 1absent - - 1--------------------- ----------- ------------ -----total disks 47 32 17

The following example command displays details about the disks that belong to the archive tier. In this example, the file system contains SAS disks:

# storage show tier archive

Archive tier details:



----------- -------- ----- --------- -----------

(available) 2.1-2.16 16 465.7 GiB

(available) 3.1-3.16 16 931.5 GiB

----------- -------- ----- --------- -----------


Storage addable disks:

Disk Disks Count Disk Enclosure Shelf Capacity Additional

Type Size Model License Needed Information

--------- -------- ----- --------- --------- -------------- -----------

(unknown) 4.1-4.15 15 465.7 GiB ES20 N/A

--------- -------- ----- --------- --------- -------------- -----------

Shelf Capacity License:

License Model Total Used Remaining

---------------- ----- -------- ------- ---------


CAPACITY-ARCHIVE ES30 21.8 TiB 0.0 TiB 21.8 TiB---------------- ----- -------- ------- ---------

When all disks are operating normally, the disk group totals (displayed below the list of storage details by disk group) are similar to the following example:

1 disk group working normally.1 disk group is degraded

When some disk groups are queued for or undergoing reconstruction, the disk group totals (displayed below the list of storage details by disk group) are similar to the following example:

Diskgroup Disk Progress (%) Remaining (minutes)Reconstruction

--------- -------------- ------------ -------------------dg4 4.12 40 125dg4 (pending) - ---------- -------------- ------------ -------------------1 disk is under resconstruction.1 disk is pending reconstruction.


The following example command displays details about the disks that belong to the archive tier. In this example, the file system contains LUNs in a gateway system:

# storage show tier archive

Disk Dev Additional StatusGroup----------- -------------------------------------- -----------------dg1 1-16dg2 17-26(available) 27-33----------- -------------------------------------- -----------------All 33 devs working normally.

Spindle Group Dev------------- --------------------1 1-102 11-153 16-18...16 100-102------------- --------------------

The following example shows steps to configure the file system in a single-node DD Archiver system:

1. Add all the available disks in enclosure 2 to the active tier, and add all the available disks in enclosures 3 and 4 to the archive tier:

# storage add tier active enclosure 2 # storage add tier archive enclosure 3 # storage add tier archive enclosure 4

2. Provision storage to the archive unit:

# filesys archive unit add# filesys archive unit expand archive-unit-1

3. List information available so far about the archive unit:

# filesys archive unit list


4. Destroy the file system, and then create a new file system with the storage in the active tier:

# filesys destroy # filesys create
