StorageGRID Webscale 10.1 Grid Designer Guide

StorageGRID® Webscale 10.1

Grid Designer User Guide

NetApp, Inc. 495 East Java Drive Sunnyvale, CA 94089 U.S.

Telephone: +1 (408) 822-6000 Fax: +1 (408) 822-4501 Support telephone: +1 (888) 463-8277 Web: www.netapp.com Feedback: [email protected]

Part number: 215-09396_A0 April 2015

http://www.netapp.com/

mailto:[email protected]

Contents1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

About this guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Who should read this guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

About Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5What you can do with Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Workflow for using Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7About grid specification files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 Installing and updating Grid Designer . . . . . . . . . . . 9Planning your deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Compatibility requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Downloading Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Installing and starting Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Viewing Grid Designer version information . . . . . . . . . . . . . . . . . . . . . . . . 11Updating Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Uninstalling Grid Designer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3 Deployment topologies and rules . . . . . . . . . . . . . . . 13Deployment topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Data center sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Grid node rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Admin Node rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14API Gateway Node rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Storage Node rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Archive Node rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Naming conventions for grid nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

4 Creating or modifying a deployment . . . . . . . . . . . . 17Creating a new StorageGRID Webscale system . . . . . . . . . . . . . . . . . . . . . 17

Saving a partially completed system . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Entering the grid ID license text block . . . . . . . . . . . . . . . . . . . . . . . . . . 19Entering password information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Entering grid node information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Entering grid network information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22Entering supplementary network information . . . . . . . . . . . . . . . . . . . 24Entering time synchronization information . . . . . . . . . . . . . . . . . . . . . 25Generating the grid specification file . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Using the grid specification file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27Modifying a StorageGRID Webscale configuration . . . . . . . . . . . . . . . . . . 27

215-09396_A0 StorageGRID Webscale 10.1

Grid Designer User Guide | 4

Obtaining the provisioned grid specification file . . . . . . . . . . . . . . . . . 28

5 Expanding an existing deployment . . . . . . . . . . . . . 31Supported expansions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Expanding a StorageGRID Webscale system . . . . . . . . . . . . . . . . . . . . . . . . 31

6 Maintenance procedures . . . . . . . . . . . . . . . . . . . . . 33Changing IP addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33Decommissioning grid nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Copyright information . . . . . . . . . . . . . . . . . . . . . . . 47

Trademark information . . . . . . . . . . . . . . . . . . . . . . 49

How to send comments . . . . . . . . . . . . . . . . . . . . . 51

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

StorageGRID Webscale 10.1 215-09396_A0

1
Introduction
About this guide

The Grid Designer User Guide describes how to use Grid Designer and create the configuration materials required to install, expand, and maintain a StorageGRID Webscale system. This guide does not include design information needed to plan out a grid topology.

Who should read this guideThis guide is intended for technical personnel trained to install, expand, maintain, and support the StorageGRID Webscale system.

About Grid Designer

Grid Designer is a Microsoft Windows application used to create the configuration materials needed to install, expand, and maintain a grid. It produces a grid specification file containing the grid’s configuration details (in an XML format) required for the successful deployment of a StorageGRID Webscale system.


6 | Grid Designer User Guide

Figure 1: A New StorageGRID Webscale System is Created

What you can do with Grid DesignerUse Grid Designer to accomplish the following tasks: • Create a new StorageGRID Webscale system.

See Chapter 4: “Creating or modifying a deployment.”• Expand a StorageGRID Webscale system.

See Chapter 5: “Expanding an existing deployment.”• Maintain a StorageGRID Webscale system, for example, by

changing an IP address or decommissioning a grid node. See Chapter 6: “Maintenance procedures.”


Introduction | 7

Workflow for using Grid DesignerUsing Grid Designer helps you create, install, expand, and maintain a StorageGRID Webscale system. Figure 2 shows how Grid Designer fits into the overall process for these operations.

Figure 2: Grid Designer Workflow

About grid specification filesAfter entering topology, server, and networking details into Grid Designer, Grid Designer generates a grid specification file.

The grid specification file is an XML file that encapsulates the configu-ration information needed to install, expand, and maintain a StorageGRID Webscale system. The file includes topology, servers, options, and networking details for the StorageGRID Webscale system.

Expansion

Create expansion with Grid Designer

Expansion grid spec file

EXP_GID<Grid_ID>_REV2_GSPEC.xmlContains new servers

Deploy grid with Grid Designer

Grid spec file

GID<Grid_ID>_REV2_GSPEC.xmlContains customer-specific information

Install expansion

Provisioned grid spec file

GID<Grid_ID>_REV2_GSPEC.xmlFile has been modified by provisioning

Create new grid with Grid Designer

Deploy grid with Grid Designer

New StorageGRID System

Grid spec file

<filename>.xmlGrid spec file contains the basic design

Grid spec file

GID<Grid_ID>_REV1_GSPEC.xmlCustomer-specific information such as

IP addresses has been added

Install StorageGRID system



Maintenance

Grid spec file

GID<Grid_ID>_REV2_GSPEC.xmlContains new information

Perform maintenance procedure



Perform maintenance procedures with Grid Designer

Grid ID

signed-grid-id.txt



Grid specification file status

The grid specification file has the following statuses: • Grid specification file: After you enter details in Grid Designer,

you generate the grid specification file from Grid Designer. • Provisioned grid specification file: You use the grid specification

file during the StorageGRID Webscale system installation and pro-visioning. After you provision the StorageGRID Webscale system, you can export a grid specification file, but it is now a provisioned grid specification file.

Grid specification file naming convention

Grid specification files use the naming convention GID<grid_ID>_REV<revision_number>_GSPEC.xml, where <grid_ID> refers to the StorageGRID Webscale system’s unique identifier and <revision_number> refers to the revision number of the grid specifica-tion file, for example, GID1234_REV1_GSPEC.xml.

The revision number is increased by one each time the grid specifica-tion file is modified, for example to add grid nodes or change IP addresses. For the initial installation of the StorageGRID Webscale system, the revision number is 1 (REV1) — that is, the grid specifica-tion file has been modified once for the installation of the StorageGRID Webscale system. Any other revision number causes provisioning to fail.


2
Installing and updating Grid Designer
Planning your deployment

Before you install Grid Designer, you should be aware of the following: • Compatibility requirements• Prerequisite software, storage, and hardware

Compatibility requirementsGrid Designer is closely linked to a specific version of both the Stor-ageGRID Webscale software and grid specification file. If you work on different releases of StorageGRID Webscale software, install the correct versions of Grid Designer to support the StorageGRID Webscale software and grid specification files on which you are working.

StorageGRID Webscale software

Each major version of Grid Designer supports a specific major release of the StorageGRID Webscale software. For example, Grid Designer 10.0 is compatible only with StorageGRID Webscale 10.0.x software. Updates to a major release of Grid Designer do not affect compatibility with the same major StorageGRID Webscale software release.

Grid specification files

Grid Designer supports and opens only grid specification files created by the same version of Grid Designer.

PrerequisitesGrid Designer requires the following items: • Computer running Windows 7



• 260 MB free disk space for Grid Designer and the .NET Framework • Microsoft .NET Framework 3.5 or 4.0

http://www.microsoft.com/net/Download.aspx

Downloading Grid Designer

You can download the Grid Designer software from the NetApp Support Site.

Steps

1. Access the NetApp Support Site at mysupport.netapp.com. 2. At the top of the page, click Downloads.3. Click Software. 4. From the StorageGRID Webscale row, select All Platforms in the

drop-down list and click Go.5. Continue to the Download page and, under the Additional

software section, download the Grid Designer software.

Installing and starting Grid Designer

The Grid Designer software is provided as a self-extracting zipped file.

1. Double-click the self extraction Grid Designer installer and follow its prompts. By default, Grid Designer installs at C:\Program Files (x86)\NetApp\

StorageGRID Webscale Grid Designer <version>, and shortcuts appear in the Windows Start menu and on the desktop. Install each version of Grid Designer in its own directory.

2. Start Grid Designer:• Double-click the Grid Designer desktop shortcut.

— or —• From the Windows Start menu, select Programs NetApp

StorageGRID Webscale Grid Designer <version>.


http://mysupport.netapp.com

http://www.microsoft.com/net/Download.aspx

Installing and updating Grid Designer | 11

Viewing Grid Designer version information

You can identify the version of Grid Designer.

1. Select About About Grid Designer.

Updating Grid Designer

Updates of Grid Designer software might be issued after its initial release. Install the latest update of Grid Designer to access its latest features. Updates might include support for new hardware and new functionality.

When you install the latest update of Grid Designer, it replaces any previous updates of that release. Compatible grid specification files are left unchanged and remain available for use with the latest update of Grid Designer.

Steps

1. Double-click the self-extracting installer for the update of Grid Designer software.

2. Follow the prompts.

Uninstalling Grid Designer

If you no longer want to use Grid Designer, you can uninstall it.

1. From the Windows Start menu, select Programs NetApp Stor-

ageGRID Webscale Grid Designer <version> Uninstall NetApp

Grid Designer <version>.2. Follow the prompts in the Uninstall wizard.


3
Deployment topologies and rulesSupported deployment topologies, rules on how grid nodes can be used, and naming conventions
Deployment topologies

Grid Designer supports the following: • One to 16 data center sites• Up to 300 grid nodes per StorageGRID Webscale deployment

Data center sites

You can create a StorageGRID Webscale deployment with a single data center or multiple data centers.

Single data center site

A single data center site (DC) topology is configured to centralize all StorageGRID Webscale infrastructure and operations in a single site. Storage can include a combination of disk storage, nearline archival storage, and cloud archival storage.

Multiple data center sites

A multiple data center site topology is configured to spread Storage-GRID Webscale infrastructure and operations across two or more data centers. Data sharing and disaster recovery is achieved in a peer-to-peer manner by automatically distributing data to other sites. Each data center (DC) site acts as a disaster recovery site for another site. Storage at any site can include a combination of disk storage, nearline archival storage, and cloud archival storage.

For more information about topologies, see the Grid Primer.



Grid node rules

When you create a StorageGRID Webscale deployment, there are unique issues that should be considered for each grid node type.

For more information about grid nodes, see the Administrator Guide.

Admin Node rulesEach StorageGRID Webscale deployment includes a primary Admin Node that monitors system operations and performs configu-ration tasks. Only the primary Admin Node includes the Configuration Management Node (CMN) service.

The following deployment rules apply to Admin Nodes:• All deployments include one primary Admin Node.• Each site should include an Admin Node; however, this is not a

mandatory configuration.• Only one Admin Node is permitted per site.

API Gateway Node rulesAPI Gateway Nodes are optional. They provide load balancing func-tionality to the StorageGRID Webscale system and connect directly to client applications. The API Gateway Node helps distribute workloads when multiple client applications are performing ingest and retrieval operations to the same data center site.

Storage Node rulesStorage Nodes host disk storage and manage objects, both object metadata and object data. All deployments must include Storage Nodes. The number of Storage Nodes depends on object storage requirements.

The following deployment rules apply to Storage Nodes:• Each data center site must include at least three Storage Nodes. • The first three Storage Nodes per site include the Administrative

Domain Controller (ADC) service.


Deployment topologies and rules | 15

• Consider proposed ILM rules when deploying Storage Nodes: the number of copies that will be made, whether replicated copies, erasure coded copies, or both will be created, and the amount of storage that is expected to be consumed.The number of Storage Nodes installed directly correlates to avail-able erasure code schemes. There is a one to one relationship between Storage Nodes and erasure code fragments. For example, to use a 6+3 erasure code scheme, there must be at least nine Storage Nodes installed with the StorageGRID Webscale system.

Archive Node rulesArchive Nodes are optional. Archive Nodes can be configured to archive object data to Amazon Simple Storage Service (S3) or to tape through a Tivoli Storage Manager (TSM) server.

Erasure coded copies of object data cannot be archived.

Naming conventions for grid nodes

Grid Designer assigns each grid node a name that uniquely identifies its location and purpose. The elements that make up a grid node name help to identify the type and function of the grid node. Grid node names can include all or part of the elements in the following format:

<Site>n–<GridNode>n <Site>n–<Type>n

Table 1: Naming convention for grid nodes

Element Description

<Site>n Identifies the site name abbreviation. By default, the abbreviated site names are:

DC Data Center

n — Each DC site is numbered sequentially.



The following are examples of grid node names:• DC2-ADM1 — Admin Node at DC2.• DC1-S4 — Storage Node at DC1.

<GridNode>n Identifies the name of the grid node:

ADM Admin Node

ARC Archive Node

G API Gateway Node

S Storage Node

n — Each grid node type at a site is numbered sequentially.

Table 1: Naming convention for grid nodes (cont.)

Element Description


4
Creating or modifying a deployment
Creating a new StorageGRID Webscale system

To create a new StorageGRID Webscale system, complete the following tasks: • Add one or more data center sites. • Add grid nodes to data center sites. • Enter password information. • Enter network information. • Enter any supplementary network information. • Enter time synchronization information.

Steps

1. From Grid Designer, select File New.A yellow field or tab indicates that required information is missing.

Figure 3: New StorageGRID Webscale System



2. In the General tab, change the grid name and description. The name can be up to 20 characters long.In the Grid Topology tree, the name of the grid site changes.

NOTE This name also appears at the top level of the Grid Topology tree in the NMS Management Interface (MI).

3. In the Grid Topology tree, add a data center site by right-clicking on the grid name, selecting Add Site, and entering a site name.Add additional data center sites as needed.

4. (Optional) Complete the following steps to change a data center site name:a. In the Grid Topology tree, right-click on the data center site and

select Rename ‘Data Center n’.b. In the Rename Site dialog box, enter the new name in Site and

the abbreviated name in Abbreviation.c. Click OK.

5. Add grid nodes to the data center by completing the following: a. In the Grid Topology tree, right-click on the data center and

select Add Nodes to <data_center>. b. Select the grid node type (Admin, Storage, API Gateway, or

Archive) and enter how many grid nodes of that type.

NOTE You can enter only one Admin Node. You must enter a minimum of three storage nodes.

c. Click OK. d. Repeat for each grid node type you need to add.

6. Continue with “Entering the grid ID license text block” on page 19.

Saving a partially completed systemAs you create your StorageGRID Webscale system, you can save the grid specification file you are creating at any time.

Steps

1. At any time during creation, select File Save As. 2. Locate a directory for your StorageGRID Webscale system.


Creating or modifying a deployment | 19

3. In the File Name box, type a name for the partial grid, and click Save.

4. To continue working on the partial grid, select File Open. 5. Locate the grid specification file and click Open. 6. Click Action Modify New Grid to continue working.

Entering the grid ID license text blockEach StorageGRID Webscale system has a unique numeric identifier. The Grid ID field is extracted from the grid license.

You must enter the Grid ID signed text block, which provides the Stor-ageGRID Webscale system with a unique grid identifier. The Grid ID signed text block comes from the signed-grid-id.txt file issued to you.

A new Grid ID signed text block is not required for expansions.

Steps

1. Open the text file (signed-grid-id.txt) that contains the Grid ID serial number.

2. Select the text including the opening and closing delimiters and all blank lines that are a part of the file. For example:-----BEGIN PGP SIGNED MESSAGE-----

Hash: SHA1

gid=xxxxxxx

-----BEGIN PGP SIGNATURE-----

Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIRtvFUycC0gvF84oRAoXpAJ0cRfml35+eJSWt6F5/Le+H36YQzgCfUSyO

FSJiBm+KxqXNudn+89uvcPA=

=FQi3

-----END PGP SIGNATURE-----

3. Press <Ctrl>+C to copy the selected text to the clipboard.4. In the General tab, place the cursor in the Enter Grid ID text block

field, and press <Ctrl>+V to paste the content into the Grid Designer field.

5. Continue with “Entering password information” on page 20.



Entering password informationWhen you create a new StorageGRID Webscale system, you also specify the password information needed to access grid nodes once the system is deployed.

You enter information for the following types of passwords: • Root password – Identifies the root password for the grid nodes.• Server Manager password – Prevents unauthorized access to

actions performed from the Server Manager console of each grid node.

• NMS MI Vendor password – Accesses the NMS Management Interface with the vendor account.

Steps

1. In the General tab, for the root password, select how it will be generated:• Auto-Generated – Create a unique random password for each server.

Choose this option for optimal security. The root password for each grid node is randomly generated and unique.

• User-Specified – Let me specify a single password for all servers. The StorageGRID Webscale system uses the same root password for all grid nodes and permits you to specify its value. This is not recommended for StorageGRID Webscale system where grid node and data security is a concern.

• Default – Use the default server password. NOT RECOMMENDED. The StorageGRID Webscale system uses the default grid node password.

Repeat the process for the Server Manager and NMS MI Vendor passwords.

WARNING For security reasons, use of the default password is not recommended.

2. If you selected the user-specified password, enter the password information in the Enter Password and Confirm Password fields.

3. Click the Grid Node tab. 4. Continue with “Entering grid node information” on page 21.



Entering grid node informationYou can change the name of a grid node and view the type of hardware for that grid node. Grid Designer automatically manages the placement of services for the StorageGRID Webscale system.

NOTE The Hardware value for both Storage Nodes installed on virtual machines and StorageGRID Webscale appliance Storage Nodes is set to VM_Storage_Node and cannot be changed.

If required, you can:• Choose which Admin Node is the primary Admin Node. The

primary Admin Node hosts the CMN service. By default, the primary Admin Node is placed at the first data center (DC) site. To change the location of the primary Admin Node, remove the CMN service from the first data center site Admin Node and add it to another Admin Node. A StorageGRID Webscale system includes only one CMN service.

• Change which Storage Nodes host the ADC service. By default, the ADC service is placed on each of the first three Storage Nodes per site. To change this, remove the ADC service from one Storage Node and add it to another Storage Node in the same site. You must have three ADC services per site.

Steps

1. (Optional) In the Grid Nodes tab, to change the grid node name, click on the name and enter a new name.



Figure 4: Grid Nodes and their Services2. (Optional) For each grid node, to change the services on that node

click on the Services cell, select the service from the view that appears, and click Finish.

3. Continue with “Entering grid network information” on page 22.

Entering grid network informationYou can specify IP addresses and network routing for the grid nodes. You configure the network to place all grid nodes on the grid network, with all traffic between grid nodes and all interaction with outside servers occurring on this single network.

When you deploy a StorageGRID Webscale system, Grid Designer automatically assigns the grid node’s network interfaces (NICs) as you make deployment decisions. The first NIC (eth0) is assigned to the grid network. If you choose to bond this network interface, eth0 and eth1 are both assigned to the grid network.



Steps

1. In the Grid Network tab, select the grid node that you want to edit.

Figure 5: Grid Network Tab2. Complete the following for the selected grid node:

a. Bonded. Select Yes or No to assign the server’s NICs as required. By default, network interfaces are not bonded.

b. IP address. Enter the IP address for the grid node. Each grid node must have a unique IP address.

c. Mask. Enter the network mask for the subnet to which the grid node has direct access.

d. Gateway. Enter the IP address of the default gateway. 3. Continue with “Entering supplementary network information” on

page 24.

Table 2: Network Interfaces

Interface NIC Server

Bonded eth0 All

Not bonded eth0, eth1 All



Entering supplementary network informationUsing the Supplementary Network tab, you can optionally specify routing for servers that must be integrated on networks other than the grid networks. You can ignore this tab if the StorageGRID Webscale system is not accessed from other networks.

Steps

1. In the Supplementary Network tab, select the Supplementary Routing Profile row that you want to edit and enter the following information:

Figure 6: Supplementary Network Taba. Profile Name. Enter a routing profile name for each route

required to configure the networking for the grid nodes. The name for the routing profile is for your reference only. On expansion, existing profiles cannot be modified. When you change the IP address of an existing network, you can edit the current profile or add a new one.

b. Network ID and Mask. Enter the Network ID and subnet Mask to specify the range of IP addresses that are routable through the network Gateway specified for the profile.

c. Gateway. Enter the IP address of the gateway that routes the addresses described by the Network ID and Mask.



2. Select the Supplementary Grid Networking row that you want to edit and enter the following information: a. Grid Node. The list contains all servers in the grid that have

enough NICs for the specified networking configuration. b. Bonded. Select Yes or No to assign bonding.

By default, network interfaces are not bonded. c. IP Address. The IP address of the grid node on the supplemen-

tary network. Each grid node must have a unique IP address.d. Mask. The network mask for the local subnet.e. Supplementary Routing Profile Name. Click to display the

Routing Profile Selection dialog, which contains the list of routing profiles that you created at the top of this tab.Select one or more routing profiles for the grid node, and then click Finish. All of the routing profiles that you selected are now assigned to the grid node.If you later expand a StorageGRID Webscale system, you can add new routing profiles to an existing grid node, but you cannot remove existing profiles.

3. Continue with “Entering time synchronization information” on page 25.

Entering time synchronization informationYou should specify the NTP configuration for the StorageGRID Webscale system. Specify at least four external NTP sources. The maximum is eight. External time sources must use the NTP protocol, not the SNTP protocol. In particular, the Windows Time Service does not provide enough synchronization accuracy because it uses SNTP.

If you later expand a StorageGRID Webscale system, the NTP sources and roles of existing grid nodes cannot be modified.

Steps

1. In the Time Synchronization tab, select the External NTP Sources row that you want to edit and enter the following information: a. Name. This field is included for reference only. Grid Designer

does not perform a DNS lookup or save this information. Enter the name of an NTP time reference that is external to the Stor-ageGRID Webscale system.



b. IP Address. The IP address of the NTP reference time source external to the StorageGRID Webscale system. Specify four external NTP time references of Stratum 3 or better to prevent issues with time drift.The IP address identifies and contacts the NTP time server, so it must be specified correctly.

Figure 7: Time Synchronization Tab2. Under Internal NTP Hierarchy, select a grid node and enter the fol-

lowing information: a. Grid Node. This field is prepopulated with the hostname of a

grid node. With the recommended NTP hierarchy, Admin Nodes and API Gateway nodes receive time and act as primary NTP time servers for the StorageGRID Webscale system, while all other grid nodes are NTP clients.

b. NTP Role. Specify whether the grid node acts as a primary NTP time server for the StorageGRID Webscale system or as an NTP client. The primary NTP time servers synchronize to external time sources and with each other. You must specify at least two primary NTP servers for redundancy and in case of loss of inter-site communication.In a StorageGRID Webscale system with multiple data center sites, at least one grid node at each site must be a primary NTP



time server. The primary Admin Node must be a primary NTP time server.

3. Continue with “Generating the grid specification file” on page 27.

Generating the grid specification fileAfter you click Finish, Grid Designer verifies data and generates a grid specification file.

1. Review each tab carefully.

The Finish button becomes available only when enough information is speci-fied to create an operational Storage-GRID Webscale system.

2. Click Finish.

A new grid specification file is created that can be used to provi-sion the StorageGRID Webscale system.A warning appears if Grid Designer detects duplicate IP addresses. The tabs highlighted in yellow contain these duplicate IP addresses.

NOTE Do not rename the grid specification file. The grid specification file must follow the GID<Grid_ID>_REV1_GSPEC.xml naming format. When the grid specification file does not follow the naming formation, provisioning fails during installation of the StorageGRID Webscale software.

Using the grid specification file

The grid specification file generated by Grid Designer is used to provi-sion the StorageGRID Webscale system. For more information about provisioning the StorageGRID Webscale system, see the Installation Guide.

Modifying a StorageGRID Webscale configuration

You can modify a grid specification file at any time before you use it to provision a StorageGRID Webscale system. You can modify it by opening it in Grid Designer, making changes, and then clicking Finish.



However, once a grid specification file is used to provision a Storage-GRID Webscale system, you must use the provisioned grid specification file to make changes. You cannot use any other version of the grid specification file. The provisioned grid specification file is used to expand the StorageGRID Webscale system or to perform main-tenance procedures.

For more information, see “Obtaining the provisioned grid specifica-tion file” on page 28, “Expanding an existing deployment” on page 31 or “Maintenance procedures” on page 33.

Steps

1. Select File Open. Locate the grid specification file (not provi-sioned), and click Open.

2. Select Actions Modify New Grid.3. Modify the StorageGRID Webscale design as required.4. Click Finish, and save the grid specification file.

The grid specification file is saved under the same name, if the Grid ID serial number has not been modified (the revision number is not updated), and under a different name, if the Grid ID serial number has been modified, following the file naming convention GID<Grid_ID>_REV1_GSPEC.xml.

Obtaining the provisioned grid specification fileBefore you edit the grid specification file, you should obtain a copy of the provisioned grid specification file from the StorageGRID Webscale system. Then, you can open the latest grid specification file in Grid Designer to edit it.

To expand a StorageGRID Webscale deployment or perform certain maintenance procedures, you must use the latest version of the provi-sion grid specification file. Obtain this file by copying it from the StorageGRID Webscale deployment.

Prerequisites• Passwords.txt file for the grid• Service laptop• An SCP tool, such as WinSCP (available at http://winscp.net/eng/

download.php), to transfer files to and from the primary Admin Node


http://winscp.net/eng/download.php

http://winscp.net/eng/download.php


Steps

1. At the primary Admin Node server, access a command shell and log in as root using the password listed in the Passwords.txt file.

2. Create a directory to which you will copy the latest provisioned grid specification file by entering: mkdir <directory>

3. Copy the latest grid specification file to the directory that you created by entering: copy-grid-spec <directory>

4. Use WinSCP or a similar tool to copy the GID<Grid_ID>_REV<revi-sion_number>_GSPEC.xml file from the newly created directory to your service laptop.

5. Log out of the command shell by entering: exit


5
Expanding an existing deploymentProcedures to expand a StorageGRID Webscale system
Supported expansions

Grid Designer supports the following expansion procedures:• Add data center sites — Add new data center sites. Each new data

center site must include at least three Storage Nodes.• Add storage capacity — Add storage capacity by adding

Storage Nodes or an Archive Node.• Add other grid nodes — Add API Gateway Nodes.

The grid node rules apply to grid expansions. For more information, see Chapter 3, “Deployment topologies and rules”.

Expanding a StorageGRID Webscale system

You can add a new data center site to a deployment or grid nodes to an existing data center site at any time.

Prerequisites• A copy of the latest provisioned grid specification file.

To expand a StorageGRID Webscale deployment or perform main-tenance procedures that require you to provision the StorageGRID Webscale system, you must use the latest version of the provision grid specification file. Obtain this file by exporting it from the Stor-ageGRID Webscale system. For more information, see “Obtaining the provisioned grid specifi-cation file” on page 28.

• A version of Grid Designer compatible with the provisioned grid specification file



• The IP address of each new grid node• Expansion Guide

Steps

1. From Grid Designer, select File Open.2. Open the latest provisioned grid specification file. 3. Select Actions Expand Grid.

The Grid Topology tree displays the sites that are already included in the grid.

4. Add data center sites and grid nodes using the same steps as for a new StorageGRID Webscale system. See “Creating a new Storage-GRID Webscale system” on page 17.

5. Update information in all applicable tabs.A yellow field or tab indicates that required information is missing.

6. When you complete the expansion, click Finish.Grid Designer creates files in same directory as the provisioned grid specification file that was used to begin the expansion process.

The next step is to expand your StorageGRID Webscale deployment using the procedures described in the Expansion Guide.


6
Maintenance proceduresMaintenance procedures that update the provisioned grid specification file
Changing IP addresses

The following IP addresses can be changed at any time:• External NTP sources IP address• Grid network IP address• Supplementary IP address

This procedure is part of a larger procedure described in the Adminis-trator Guide.

Prerequisites• A copy of the latest provisioned grid specification file

To perform certain maintenance procedures, you must use the latest version of the provisioned grid specification file. Obtain this file from the StorageGRID Webscale system. For more information, see “Obtaining the provisioned grid specifi-cation file” on page 28.

• A version of Grid Designer that is compatible with the currently installed version of StorageGRID Webscale software

• The new IP addresses• Administrator Guide

Steps

1. From Grid Designer, select File Open.2. Open the latest provisioned grid specification file.



3. Select Maintenance Change IP Addresses. 4. Click the Grid Network tab, the Supplementary Network tab, or the

NTP tab.

NOTE Both networks and NTP can be changed at the same time.

5. Make the required changes. You can change fields that have a white background. For a description of the fields, see “Creating a new StorageGRID Webscale system” on page 17.

6. Change values on other tabs as needed. 7. Click Finish.

The grid specification file is saved, with the revision number increased by 1.

This grid specification file can now be used to provision the Storage-GRID Webscale system.

After you finish

To complete the IP address changes, follow the instructions in the Administrator Guide.

Decommissioning grid nodes

You can update the grid specification file to indicate grid nodes to decommission. You can decommission Storage Nodes and API Gateway Nodes.

Storage Node decommissioning is the process of permanently moving content from an existing Storage Node to other Storage Nodes in such a way that system operations are not affected, and the ILM policy is satisfied. After all content has been moved to other Storage Nodes, the original Storage Node is removed from the StorageGRID Webscale system.

This procedure is part of a larger procedure described in the Mainte-nance Guide.


Maintenance procedures | 35

Prerequisites• A copy of the latest provisioned grid specification file

To perform certain maintenance procedures, you must use the latest version of the provisioned grid specification file. Obtain this file from the StorageGRID Webscale system. For more information, see “Obtaining the provisioned grid specifi-cation file” on page 28.

• A version of Grid Designer that is compatible with the currently installed version of StorageGRID Webscale software

• The name or names of the Storage Nodes to decommission• Maintenance Guide

Steps

1. From Grid Designer, select File Open.2. Open the latest grid specification file. 3. Select Maintenance Decommission a Storage or Gateway Node.4. Click to select one Storage Node or API Gateway Node to decom-

mission, or Ctrl+click to select multiple grid nodes, and click OK.

Figure 8: Selecting Storage Nodes for DecommissioningThe grid specification file is saved, and the revision number increases by one.

5. Click OK in the confirmation dialog box.

To complete the decommissioning procedure, follow the instructions in the Maintenance Guide.


GlossaryACL Access control list—Specifies what users or groups of users are

allowed to access an object and what operations are permitted, for example, read, write, and execute.

ADC service Administrative Domain Controller —The ADC service maintains topology information, provides authentication services, and responds to queries from the LDR, CMN, and CLB services. The ADC service is present on the first three Storage Nodes installed at a site.

ADE Asynchronous Distributed Environment—Proprietary development environment used as a framework for services within the StorageGRID Webscale system.

Admin Node The Admin Node provides services for the web interface, system con-figuration, and audit logs. See the “primary Admin Node” definition.

Amazon S3 Proprietary web service from Amazon for the storage and retrieval of data.

AMS service Audit Management System—The AMS service monitors and logs all audited system events and transactions to a text log file. The AMS service is present on the Admin Node.

API Application Programming Interface—A set of commands and func-tions, and their related syntax, that enable software to use the functions provided by another piece of software.

API Gateway Node

An API Gateway Node provides load balancing functionality to the StorageGRID Webscale system and is used to distribute the workload when multiple client applications are performing ingest and retrieval operations. API Gateway Nodes include a “CLB service” service.

ARC service Archive—Communicates with an external archive service to store objects in and retrieve objects from the archive. The ARC service is present on the Archive Node.

Archive Node An Archive Node manages the storage of objects to, and retrieval of objects from, external archive services. Archive Nodes can be config-ured to work with Amazon Simple Storage Service (S3) or Tivoli Storage Manager (TSM).

atom Atoms are the lowest-level component of the container data structure, and generally encode a single piece of information. (Containers are



sometimes used when interacting with the StorageGRID Webscalesys-tem through the StorageGRID API).

audit message Information about an event occurring in the StorageGRID Webscale system that is captured and logged to a file.

AutoYaST An automated version of the Linux installation and configuration tool YaST (Yet another Setup Tool), which is included as part of the SUSE Linux distribution.

BASE64 A standardized data encoding algorithm that enables 8-bit data to be converted into a format that uses a smaller character set, enabling it to safely pass through legacy systems that can only process basic (low order) ASCII text excluding control characters. See RFC 2045 for more details.

binding The persistent assignment of a service (for example, an AMS service or SSM service) to the NMS service. This assignment is based on Storage-GRID Webscale topology. See also the “Admin Node” definition.

bundle A structured collection of configuration information used internally by various components of the StorageGRID Webscale system. Bundles are structured in container format.

Cassandra An open-source database that is scalable and distributed, provides high availability, and handles large amounts of data across multiple servers.

CBID Content Block Identifier — A unique internal identifier of a piece of content within the StorageGRID Webscale system.

CDMI Cloud Data Management Interface — An industry standard defined by SNIA that includes a RESTful interface for object storage. For more information, see www.snia.org/cdmi.

CIDR Classless Inter-Domain Routing—A notation used to compactly describe a subnet mask used to define a range of IP addresses. In CIDR notation, the subnet mask is expressed as an IP address in dotted decimal notation, followed by a slash and the number of bits in the subnet. For example, 192.0.2.0/24.

CLB service Connection Load Balancer—The CLB service provides a gateway into the StorageGRID Webscale system for client applications connecting through HTTP. The CLB service is part of the API Gateway Node.

Cloud Data Management

Interface

See “CDMI” on page 38.


http://www.snia.org/cdmi

http://www.snia.org/cdmi

Glossary | 39

CMN service Configuration Management Node— The CMN service manages system-wide configuration and grid tasks. The CMN service is present on the primary Admin Node.

CMS service Content Management System—The CMS service carries out the opera-tions of the active ILM policy’s ILM rules, determining how object data is protected over time. The CMS service is present on the Storage Node.

command In HTTP, an instruction in the request header such as GET, HEAD,

DELETE, OPTIONS, POST, or PUT. Also known as an HTTP method.

container Within the StorageGRID Webscale system, there are several different uses (and technologies) for the term container:• SGAPI — A container is a data structure used by the internals of

StorageGRID Webscale software. In the StorageGRID API, an XML representation of a container is used to define queries or audit messages submitted using the POST command. Containers are used for information that has hierarchical relationships between components. The lowest-level component of a container is an atom. Containers can contain 0 to N atoms, and 0 to N other containers.

• CDMI — Analogous to a file system directory, a container object allows access to “child” data objects.

• segmented objects — When an object is split into segments, a con-tainer is created that lists the header information for all segments of the split object. This is then used by the LDR service to assemble the segmented object when it is retrieved by a client application.

content block ID

See the“CBID” definition.

content handle See the “UUID” definition.

CSTR Null-terminated, variable length string.

data fragment An erasure coded chunk of object data. See the “erasure code” definition.

DC Data Center site.

DDS service Distributed Data Store — The DDS service interfaces with the distrib-uted key-value store and manages object metadata. It distributes metadata copies to multiple instances of the distributed key-value store so that metadata is always protected against loss.



distributed key value store

Data storage and retrieval that unlike a traditional relational database manages data across grid nodes.

Enablement Layer

The Enablement Layer is used during installation to customize the Linux operating system installed on each grid node. Only the packages needed to support the services hosted on the grid node are retained, which minimizes the overall footprint occupied by the oper-ating system and maximize the security of each grid node.

erasure code A object data protection scheme that parses object data into data and parity fragments. See also the “data fragment” and “parity fragment” definitions.

fibre channel A networking technology primarily used for storage.

GDU Grid Deployment Utility—A StorageGRID Webscale software utility used to facilitate the installation of software on all grid nodes. GDU is installed and available on the primary Admin Node.

GPT Grid Provisioning Tool— A software tool included with StorageGRID Webscale software that permits you to provision a StorageGRID Webscale system for installation, maintenance, or expansion. GPT creates and maintains an encrypted repository of information about the system that is required to maintain the StorageGRID Webscale system and recover failed grid nodes.

Grid ID signed text block

A BASE64 encoded block of cryptographically signed data that contains the grid ID which must match the grid ID (gid) element in the grid specification file. See also “provisioning”.

grid node The basic software building block for the StorageGRID Webscale system, for example, Admin Node or Storage Node. Each grid node type consists of a set of services that perform a specialized set of tasks.

grid specification

file

An XML file that provides a complete technical description of a specific StorageGRID Webscale deployment. It describes the system’s topology and specifies the hardware, options, grid nodes, network settings and time synchronization for the deployment.

grid task System-wide scripts used to trigger various actions that implement specific changes to the StorageGRID Webscale system. For example, most maintenance and expansion procedures involve running grid tasks. Grid tasks are typically long-term operations that span many entities within the StorageGRID Webscale system. See also the “Task Signed Text Block” definition.


Glossary | 41

HTTP Hyper-Text Transfer Protocol—A simple, text based client/ protocol for requesting hypertext documents from a server.

HTTPS Hyper-Text Transfer Protocol, Secure—URIs that include HTTPS indicate that the transaction must use HTTP with an additional encryption/authentication layer and often, a different default port number. The encryption layer is usually provided by SSL or TLS. HTTPS is widely used on the internet for secure communications.

ILM Information Lifecycle Management—A process of managing content storage location and duration based on content value, cost of storage, performance access, regulatory compliance and other such factors. See also the “Admin Node” and “storage pool” definitions.

LAN Local Area Network—A network of interconnected computers that is restricted to a small area, such as a building or campus. A LAN can be considered a node to the Internet or other wide area network. Contrast with WAN.

latency Time duration for processing a transaction or transmitting a unit of data from end to end. When evaluating system performance, both throughput and latency need to be considered. See the “throughput” definition.

LDR service Local Distribution Router —The LDR service manages the storage and transfer of content within the StorageGRID Webscale system. The LDR service is present on the Storage Node.

LUN See the “object store” definition.

metadata Information related to or describing an object stored in the Storage-GRID Webscale system; for example, ingest time.

namespace A set whose elements are unique names. There is no guarantee that a name in one namespace is not repeated in a different namespace.

nearline A term describing data storage that is neither “online” (implying that it is instantly available like spinning disk) nor “offline” (which could include offsite storage media). An example of a nearline data storage location is a tape that is loaded in a tape library, but is not necessarily mounted.

NFS Network File System—A protocol (developed by SUN Microsystems) that enables access to network files as if they were on local disks.

NMS service Network Management System—The NMS service provides a web-based interface for managing and monitoring the StorageGRID



Webscale system. The NMS service is present on the Admin Node. See also “NMS MI” and “Admin Node” definitions.

NMS MI NMS Management Interface—The web-based interface for managing and monitoring the StorageGRID Webscale system provided by the NMS software component. See also “NMS service” definition.

node ID An identification number assigned to a service within the Storage-GRID Webscale system. Each service (such as an NMS service or ADC service) must have a unique node ID. The number is set during system configuration and tied to authentication certificates.

NTP Network Time Protocol—A protocol used to synchronize distributed clocks over a variable latency network such as the internet.

object An artificial construct used to describe a system that divides content into data and metadata.

object storage Object Storage is an approach where stored data is accessed by unique identifiers rather than by a user-defined hierarchy of directories and files. Each object has both data (for example, a picture) and metadata (for example, the date the picture was taken). Object storage opera-tions typically act on entire objects as opposed to reading and writing bytes as is commonly done with files, and is typically provided via APIs or HTTP instead of NAS (CIFS/NFS) or block protocols (iSCSI/FC/FCOE).

object store A configured file system on a disk volume. The configuration includes a specific directory structure and resources initialized at system installation.

object segmentation

A StorageGRID Webscale process that splits a large object into a collec-tion of small objects (segments) and creates a segment container to track the collection. The segment container contains the UUID for the collection of small objects as well as the header information for each small object in the collection. All of the small objects in the collection are the same size. See also the “segment container” definition.

OID Object Identifier—The unique identifier of an object.

parity fragment An erasure coded chunk containing information that can be used to reconstruct object data if data fragments are lost. See the “erasure code” definition.

primary Admin Node

Admin Node that hosts the CMN service. There is one per Storage-GRID Webscale system. See also the “Admin Node” definition.


Glossary | 43

provisioning The process of generating a SAID package and GPT repository. This is done on the primary Admin Node using the provision command. The new or updated SAID package is saved to the Provisioning Media. See also the “grid specification file” and “SAID” definitions.

purge The act of permanently removing an object from the StorageGRID Webscale system.

quorum A simple majority: 50% + 1. Some system functionality requires a quorum of the total number of a particular service type.

SAID Software Activation and Integration Data—Generated during provi-sioning, the SAID package contains site-specific files and software needed to install a StorageGRID Webscale system.

SATA Serial Advanced Technology Attachment—A connection technology used to connect server and storage devices.

SCSI Small Computer System Interface— A connection technology used to connect servers and peripheral devices such as storage systems.

segment container

An object created by the StorageGRID Webscale system during the segmentation process. Object segmentation splits a large object into a collection of small objects (segments) and creates a segment container to track the collection. A segment container contains the UUID for the collection of segmented objects as well as the header information for each segment in the collection. When assembled, the collection of segments creates the original object. See also the “object segmentation” definitions.

server Used when specifically referring to hardware. May also refer to a virtual machine.

Server Manager Application that runs on all grid nodes, supervises the starting and stopping of services, and monitors all services on the grid node.

service A unit of the StorageGRID Webscale system such as the ADC service, NMS service, or SSM service. Each service performs unique tasks critical to the normal operations of a StorageGRID Webscale system.

SGAPI StorageGRID Application Programming Interface—A set of commands and functions, and their related syntax, that provides client applications with the ability to connect directly to an LDR or CLB service.



SLES SUSE Linux Enterprise Server —A commercial distribution of the SUSE Linux operating system, used with the StorageGRID Webscale system.

SQL Structured Query Language— An industry standard interface language for managing relational databases. An SQL database is one that supports the SQL interface.

ssh Secure Shell— A Unix shell program and supporting protocols used to log in to a remote computer and execute commands over an authenti-cated and encrypted channel.

SSM Server Status Monitor—A component of the StorageGRID Webscale software that monitors hardware conditions and reports to the NMS service. Every grid node runs an instance of the SSM service.

SSL Secure Socket Layer—The original cryptographic protocol used to enable secure communications over the internet. See also the “TLS” definition.

Storage Node The Storage Node provides storage capacity and services to store, move, verify, and retrieve objects stored on disks.

storage pool The element of an ILM rule that determines the location where an object is stored.

StorageGRID Webscale

A registered trademark of NetApp Inc. for their object storage grid architecture and software system.

StorageGRID API

See the“SGAPI” definition.

storage volume See the “object store” definition.

Task Signed Text Block

A BASE64 encoded block of cryptographically signed data that provides the set of instructions that define a grid task.

TCP/IP Transmission Control Protocol / Internet Protocol—A process of encapsulating and transmitting packet data over a network. It includes positive acknowledgment of transmissions.

throughput The amount of data that can be transmitted or the number of transac-tions that can be processed by a system or subsystem in a given period of time. See also the “latency” definition.

TLS Transport Layer Security—A cryptographic protocol used to enable secure communications over the internet. See RFC 2246 for more details.


Glossary | 45

transfer syntax The parameters, such as the byte order and compression method, needed to exchange data between systems.

Tivoli® Storage Manager

IBM storage middleware product that manages storage and retrieval of data from removable storage resources.

URI Universal Resource Identifier—A generic set of all names or addresses used to refer to resources that can be served from a computer system. These addresses are represented as short text strings.

UTC A language-independent international abbreviation, UTC is neither English nor French. It means both “Coordinated Universal Time” and “Temps Universel Coordonné.” UTC refers to the standard time common to every place in the world.

UUID Universally Unique Identifier—Unique identifier for each piece of content in the StorageGRID Webscale system. UUIDs provide client applications with a content handle that permits them to access content in a way that does not interfere with the StorageGRID Webscale system’s management of that same content. A 128-bit number which is guaranteed to be unique. See RFC 4122 for more details.

VM Virtual Machine—A software platform that enables the installation of an operating system and software, substituting for a physical server and permitting the sharing of physical server resources amongst several virtual “servers.”

XFS A scalable, high performance journaled file system originally devel-oped by Silicon Graphics.

WAN Wide Area Network—A network of interconnected computers that covers a large geographic area such as a country. Contrast with the “LAN” definition.

XML eXtensible Markup Language—A text format for the extensible repre-sentation of structured information; classified by type and managed like a database. XML has the advantages of being verifiable, human readable, and easily interchangeable between different systems.


Copyright informationCopyright © 1994–2015 NetApp, Inc. All rights reserved. Printed in the U.S.

No part of this document covered by copyright may be reproduced in any form or by any means— graphic, electronic, or mechanical, includ-ing photocopying, recording, taping, or storage in an electronic retrieval system—without prior written permission of the copyright owner.

Software derived from copyrighted NetApp material is subject to the following license and disclaimer:

THIS SOFTWARE IS PROVIDED BY NETAPP "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEM-PLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SER-VICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

NetApp reserves the right to change any products described herein at any time, and without notice. NetApp assumes no responsibility or lia-bility arising from the use of products described herein, except as expressly agreed to in writing by NetApp. The use or purchase of this product does not convey a license under any patent rights, trademark rights, or any other intellectual property rights of NetApp.

The product described in this manual may be protected by one or more U.S. patents, foreign patents, or pending applications.

RESTRICTED RIGHTS LEGEND: Use, duplication, or disclosure by the government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.277-7103 (October 1988) and FAR 52-227-19 (June 1987).


Trademark informationNetApp, the NetApp logo, Network Appliance, the Network Appliance logo, Akorri, ApplianceWatch, ASUP, AutoSupport, BalancePoint, Bal-ancePoint Predictor, Bycast, Campaign Express, ComplianceClock, Cryptainer, CryptoShred, Data ONTAP, DataFabric, DataFort, Decru, Decru DataFort, DenseStak, Engenio, Engenio logo, E-Stack, FAServer, FastStak, FilerView, FlexCache, FlexClone, FlexPod, FlexScale, Flex-Share, FlexSuite, FlexVol, FPolicy, GetSuccessful, gFiler, Go further, faster, Imagine Virtually Anything, Lifetime Key Management, Lock-Vault, Manage ONTAP, MetroCluster, MultiStore, NearStore, NetCache, NOW (NetApp on the Web), Onaro, OnCommand, ONTAPI, OpenKey, PerformanceStak, RAID-DP, ReplicatorX, SANscreen, SANshare, SANt-ricity, SecureAdmin, SecureShare, Select, Service Builder, Shadow Tape, Simplicity, Simulate ONTAP, SnapCopy, SnapDirector, SnapDrive, Snap-Filter, SnapIntegrator, SnapLock, SnapManager, SnapMigrator, SnapMirror, SnapMover, SnapProtect, SnapRestore, Snapshot, Snap-Suite, SnapValidator, SnapVault, StorageGRID, StoreVault, the StoreVault logo, SyncMirror, Tech OnTap, The evolution of storage, Topio, vFiler, VFM, Virtual File Manager, VPolicy, WAFL, Web Filer, and XBB are trademarks or registered trademarks of NetApp, Inc. in the United States, other countries, or both.

IBM, the IBM logo, and ibm.com are trademarks or registered trade-marks of International Business Machines Corporation in the United States, other countries, or both. A complete and current list of other IBM trademarks is available on the Web at www.ibm.com/legal/copytrade.shtml.

Apple is a registered trademark and QuickTime is a trademark of Apple, Inc. in the United States and/or other countries. Microsoft is a registered trademark and Windows Media is a trademark of Microsoft Corporation in the United States and/or other countries. RealAudio, RealNetworks, RealPlayer, RealSystem, RealText, and RealVideo are registered trade-marks and RealMedia, RealProxy, and SureStream are trademarks of RealNetworks, Inc. in the United States and/or other countries.

All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such.

NetApp, Inc. is a licensee of the CompactFlash and CF Logo trademarks.

NetApp, Inc. NetCache is certified RealSystem compatible.


http://www.ibm.com/legal/us/en/copytrade.shtml

http://www.ibm.com/legal/us/en/copytrade.shtml

How to send commentsYou can help us to improve the quality of our documentation by sending us your feedback.

Your feedback is important in helping us to provide the most accurate and high-quality information. If you have suggestions for improving this document, send us your comments by email to [email protected]. To help us direct your comments to the correct division, include in the subject line the product name, version, and oper-ating system.

You can also contact us in the following ways:• NetApp, Inc., 495 East Java Drive, Sunnyvale, CA 94089 U.S.• Telephone: +1 (408) 822-6000• Fax: +1 (408) 822-4501• Support telephone: +1 (888) 463-8277




IndexAabbreviations

grid nodes 16server names 16site names 15

ACLsdefined 37

ADCdefined 37

ADEdefined 37

Admin Nodes 14defined 37HCAC 15NTP servers 26server location 14

AMSdefined 37

APIdefined 37

API Gateway Nodedefined 37

API Gateway Nodesrules 14

ARCdefined 37

Archive Nodesdefined 37rules 15

atomdefined 37

audit messagedefined 38

audit messagesdefined 38

AutoYaSTdefined 38

BBASE64

defined 38binding

defined 38bundle

defined 38

CCassandra

defined 38CBID

defined 38changing

grids 27IP addresses 33Storage Node information 34

CIDRdefined 38

CLBdefined 38

CMNdefined 39

CMS servicedefined 39

command, HTTPdefined 39

containerdefined 39

contentblock ID defined 39

content handledefined 39

CSTRdefined 39

custom expansions 31customer network

IP address changes 33network interface

assignment 22

Ddata centers

defined 39DC 13

defined 39multiple 13single 13topology 13

DDS servicedefined 39

decommissioning Storage Nodes 34

designing new grids 17disk space required for install 10distributed key value store

defined 40DR 13duplicate IP addresses 27

EEnablement Layer

defined 40erasure code

data fragment defined 39defined 40parity fragment defined 42

ethernet cards 22expansion

custom expansions 31grid specification files 31, 33,

35supported expansions 31

FFibre Channel

defined 40

GGDU

defined 40Grid Designer

about 5disk space 10installing 9uninstalling 11updating 11workflow 7

Grid IDsigned text block 19

grid nodesnames of 15naming convention 15rules 14

Grid Provisioning Tooldefined 40

grid rules 13Admin Nodes 14API Gateway Nodes 14Archive Nodes 15

grid specification fileschanging IP addresses 34changing link cost groups 34export from grid 33required for expansion 31, 33,

35Storage Node

decommissioning 35



workflow 7grid topology

DC 13overview 13

gridsID signed text block 40nodes 40specification files 40tasks 40

Hhardware

database 11refresh 34

HCAC 15HTTP

defined 41HTTPS

defined 41

IILM

defined 41installing

disk space required 10Grid Designer 9, 10Grid Designer updates 11Microsoft .NET Framework

3.5 10IP addresses

change 33customer network 33duplicate 27grid network 33NTP 33virtual IP address of cluster 33

LLAN

defined 41latency

defined 41LDR

defined 41

Mmaintenance procedures 33

export grid specification files 33

hardware refresh 34IP address changes 33Storage Node

decommissioning 34

metadatadefined 41

Microsoft .NET Framework 3.5 10

Nnamespace

defined 41naming convention 15nearline

defined 41networking

ethernet cards 22NFS

defined 41NICs 22NMS

defined 41NMS MI

defined 42node ID

defined 42NTP

client servers 26defined 42IP address changes 33primary Admin Node 27primary NTP server 26primary time servers 26recommended number of

sources 25Time Synchronization tab 25updating 26

Oobjects

defined 42segmentation defined 42storage 42stores defined 42

objectsidentifier. See OID 42

OIDdefined 42

Pprimary Admin Node

defined 42private network

change IP address 33network interface

assignment 22provisioning

defined 43purging objects

defined 43

Rremoving Grid Designer 11rules 14

SS3

defined 37SAID package

defined 43SATA

defined 43SCSI

defined 43segment container

defined 43server

Admin Nodes 14defined 43

Server Managerdefined 43

servicesdefined 43

signed-grid-id.txt file 19SLES

defined 44SQL

defined 44ssh

defined 44SSL

defined 44SSM

defined 44Storage Nodes

decommissioning 34defined 44

storage pooldefined 44

StorageGRID Webscalesoftware version 9

StorageGRID Webscale systemdefined 44

Supplementary Network tab 24

Ttask signed text block

defined 44TCP/IP

defined 44throughput

defined 44Time Synchronization tab 25


Index | 55

TLSdefined 44

topologymultiple data centers 13single data center 13

transfer syntaxdefined 45

TSMdefined 45

Uuninstalling Grid Designer 11updating Grid Designer 11URI

defined 45UTC

defined 45UUIDs

defined 45

Vversions

StorageGRID Webscale software 9

virtual IP address, change 33virtual machine

definition 45

WWAN

defined 45WinSCP 28workflow 7

XXFS

defined 45XML

defined 45

Yyellow in tab or field 17


StorageGRID Webscale 10.1 Grid Designer Guide

Documents

Transcript of StorageGRID Webscale 10.1 Grid Designer Guide