CloudByte ElastiStor Guide

CloudByte ElastiStor Patch 5

Installation andAdministration GuideBuild: 1.4.0.790

1

Copyright 2016 CloudByte Inc.. All rights reserved.

Information in this document is subject to change without notice. The software described in this document is furnished under alicense agreement or nondisclosure agreement. The softwaremay be used or copied only in accordance with the terms of thoseagreements. No part of this publicationmay be reproduced, stored in a retrieval system, or transmitted in any form or any meanselectronic or mechanical, including photocopying and recording for any purpose other than the purchaser's personal use without thewritten permission of CloudByte Inc..

CloudByte Inc.

4675 Stevens Creek Blvd, Suite 122

Santa Clara, CA 95051

USA

+1-(408)-663-6900

[email protected]

www.cloudbyte.com

2

CONTENTS

CHAPTER I Getting started with CloudByte ElastiStor 8

Introduction 8

CloudByte ElastiStor feature highlights 11

Physical storage infrastructure 14

Installation overview 17

Hardware Compatibility List 18

Host interoperability matrix 19

CHAPTER II What's new in ElastiStor 1.4.0 Patch 5 20

What's new in ElastiStor 1.4.0 Patch 4 20




What's new in CloudByte ElastiStor 1.4 40

CHAPTER III Installing CloudByte ElastiStor 44

Planning your installation 44

System requirements 45

Network configuration details 45

Installation options 46

Installing ElastiCenter 46

Installing ElastiCenter HA 54

Installing Node 55

3

Upgrading to CloudByte ElastiStor 1.4.0 Patch 5 64

release Moving your CloudByte ElastiStor setup to a new network environment 70

CHAPTER IV CloudByte ElastiStor Appliance 75

Installing ElastiCenter and ElastiStor 75

Moving your CloudByte ElastiStor Appliance (ESA) to a new network environment 84

Upgrading to CloudByte ElastiStor 1.4.0 Patch 5 88

CHAPTER V ElastiCenter, ElastiStor'smanagement console 97

Introduction 97

ElastiCenter User Interface 98

Launching CloudByte ElastiCenter 102

Authentication 102

Reset password and configure auto-alert options 103

Replacing default SSL Certificate 104

Customizing the CloudByte ElastiCenter theme 105

CHAPTER VI Configuring CloudByte ElastiStor 108

Setting up your storage 108

Creating a Site 112

Creating an HA Group 112

Adding a Node 114

Configuring Disk Array enclosure 115

Creating a Pool 120

Configuring Spare 123

Migrating a Pool 123

Pool encryption 123

4

Creating an Account 125

Delete account user 126

Provisioning a VSM 126

Configuring VSM migration 131

Migrant VSM activation 134

Provisioning storage 138

Choosing a Pool to provision VSM 142

SAS multipathing 144

Modifying QoS after provisioning VSM 147

Modifying QoS after provisioning Storage Volume 148

Modifying storage capacity 149

Configuring High Availability 150

Configuring local backup 156

Configuring remote disaster recovery 160

Upgrading a High Availability DR setup 163

Assuring Quality of Service 164

Monitoring QoS 167

QoS tips 167

CHAPTER VII Setting up data access protocols 169

Data access using iSCSI 169

Installing MPIO on Windows Server 180

Configuring Multipath on ElastiCenter 183

Configuring MPIO on Windows Server for a LUN in ElastiStor 185

Data access using CIFS 193

Data access using FC 198

5

Data access using NFS 201

Restore snapshots while mapping iSCSI LUNs 203

Restore snapshots while mapping CIFS share 210

Restore snapshots while mapping FC LUNs 214

Restore snapshots while mapping NFS Share 220

CHAPTER VIII Managing CloudByte ElastiStor 225

Delegated administration 225

Monitoring 230

Drill-down feature 234

CloudByte ElastiStor reporting 235

Global Settings 238

CHAPTER IX Licensing 242

How do I access license details? 242

CHAPTER X CloudByte ElastiStor REST APIs 245

CloudByte ElastiStor API references 245

CHAPTER XI Connecting to ElastiStor 251

Mapping FC LUN to the clients 251

Mapping iSCSI LUN 262

Resizing iSCSI LUN 266

Mapping CIFS share to the clients 274

Mapping NFS share to clients 277

CHAPTER XII Frequently asked questions 281

Product FAQs 281

Installation FAQs 284

6

Pool FAQs 287

Access Protocol FAQs 289

Storage Volume FAQs 293

QoS FAQs 294

High Availability FAQs 297

NIC FAQs 298

Global Settings 298

CloudByte ElastiStor best practices 301

Promoting secondary ElastiCenter and backing up configurations 302

IOPS conversion matrix 307

CHAPTER XIII Glossary 311

7

CHAPTER I Getting started with CloudByte ElastiStor


The following topics help you get started with CloudByte ElastiStor™:

l "Introduction"

l "CloudByte ElastiStor feature highlights"

l "Physical storage infrastructure"

l "Simple hardware setup"

l "Installation overview"

Introduction

What is CloudByte ElastiStor

CloudByte ElastiStor is a full-featured software-defined storageQoS solution, purpose-built for the cloud and virtualized envir-onments. Software-only ElastiStor makes storage predictable, affordable and easy, even as you scale to thousands of applications.

Installable on a wide range of industry-standard servers, ElastiStor frees enterprises and service providers from vendor lock-ins.ElastiStor lets you custom-build storage infrastructure based on your requirements, with support for SATA, SAS, and SSD hardwareas well as NFS, CIFS, FC, and iSCSI protocols.

Cloud-scale Performance

The allocation of application-specific storage silos to deliver performance does not scale well in today’s cloud and virtualized envir-onments. ElastiStor enables you to guarantee performance (IOPS, throughput, and latency) for every application right from a com-mon storage platform. This fully shared storage optimizes your infrastructure, minimizes your CAPEX, and enables moreperformance applications tomove to the cloud.

No Vendor Lock-in

In addition to being software-defined, ElastiStor eliminates the need for you to lock into specific vendors or overspend on proprietaryhardware. The benefits of commodity, industry-standard resources are first reflected inminimized CAPEX, and later make shiftingvendors easy with no proprietary hardware lock-in.

Hardware Flexibility

Restricting storagemedia choice is not an ideal scenario when you have a wide variety of performance needs, from archival to bigdata applications, and it is definitely not practical to fit every performance-sensitive application on an all-SSD array. ElastiStor issoftware-only so you continue to work with the storagemedia of your choice – SSD, SAS, SATA, or any mix – optimized for your per-formance needs.

Simple Management

Managing storage for a large number of applications and their constantly changing performance demands is not for the faint hearted.Manual tuning of spindles or other hardware is no longer an option. Being software-defined, ElastiStor allows you to instantly pro-vision fully-elastic storage volumes for any workload from the shared storage pool to enable fast deployment and reconfigurationoptions. Now it is a simplematter to schedule backups, migrate sites, and empower customers with comprehensivemanagementrights for their dedicated storage. Additionally, datacenter admins will be able to provision andmanageQoS-aware storage volumesfor VMs right from VMWare vCenter, Citrix XenCenter, OpenStack Horizon, or CloudStack.

8


Affordable

With CloudByte ElastiStor you can now replace hundreds of legacy storage silos in your datacenters with a single, extensible stor-age platform. By fully sharing your infrastructure, deploying industry-standard hardware and optimally choosing your storagemedia,ElastiStor can significantly reduce your datacenter footprint, leading to 80-90% cost savings over 3-5 years. Add simplifiedman-agement that brings down your operational costs, and you get the lowest cost storageQoS solution.

Features at a glance

Along with the standard storage features, CloudByte ElastiStor provides certain product-specific features of which some are unique.

CloudByte ElastiStor features

Guaranteed Storage QoS within Shared Storage

Share your storage and deliver predictable performance to every application. For the first time ever, ElastiStor allows storage end-points to be defined beyond capacity, in terms of IOPS, throughput and latency. This allows applications with diverse workloads tobe guaranteedQoS from a shared storage platform. Together with linear scaling, a single extensible shared storage platform fromElastiStor can now replace legacy solutions’ dedicated storage silos. By fully sharing storage and optimally utilizing resources,ElastiStor steeply cuts down your storage footprint, leading to 80-90% cost savings over 3-5 years.

On-Demand Performance Provisioning

Do you still manually configure hardware to provision performance for any new application? Break the need for hardwiring storagewith ElastiStor’s on-demand provisioning. Just enter the required SLA/QoS parameters and let ElastiStor automate node selectionand resource allocation for you. ElastiStor includes an intelligent heuristics daemonwhich continuously learns the quantity of vari-ous controller resources needed to deliver the required QoS.

vCenter-like Administration Console

ElastiStor makes managing storage as easy as managing VMs, even as you scale to hundreds of applications. Storage admins cannow comprehensively manage the entire storage cluster, spanning across multiple sites, from a single web-based console. Further,ElastiStor gives you unprecedented access and control over resource usage within shared storage, right down to the application-level granularity.

REST APIs and Plugins for Easy Integration

Every action performed at ElastiStor admin console translates into a REST based API call in the backend. Using these REST APIs,admins can comprehensively provision andmanage ElastiStor volumes from VMware vCenter, Citrix XenCenter, OpenStack Hori-zon, or any other admin portal. Our plugin for VMware vCenter allows admins to create andmanageQoS-aware VMs right fromvCenter console.

N-way High Availability

ElastiStor enables N-way High Availability, exponentially increasing reliability (mean time to failure), compared to the standard 2-way HA provided by existing solutions. CloudByte’s storage un-fragmentation and its patented VSM architecturemake N-way HAaffordable and feasible.

Delegated Administration

A much requested feature from the cloud service providers, delegated administration empowers both CSPs and its customers tomonitor and control storage volumes. Management privileges vary based on the admin functionality – for example, a super admin canmanage the entire storage cluster, whereas a customer admin canmanage just the storage resources allotted to that particular cus-tomer.

Standard storage features

The following Storage features are available in CloudByte ElastiStor:

9


Feature Description

Scalability l 128-bit file system

l Zettabyte storage capacity

l Unlimited file size

Access Protocols l NFSv3

l NFSv4

l CIFS

l ISCSI

l FC

Storage Connectivity l SAS JBODs

Storage Resilience l RAID Z1

l RAID Z2

Storage Efficiency l De-duplication

l Compression

l Thin Provisioning

Back up l Unlimited Clones

l Unlimited Snapshots

Availability l N-way high availability (N-way HA)

l Partial failure transfer to the available node

l Active-Active mode

l HA with/without storage redundancy

Disaster Recovery l Tenant level disaster recovery

l High availability across primary and DR sites

l Block level replication

l Synchronous mirroring,

l Asynchronous mirroring

l RPO—Last minute

l RTO—Fewminutes

Data Integrity l Protection against silent data corruption

l Fixes corrupt block without having to take the file system offline

Note: The supported capacity of ElastiStor depends on the hardware. With adequate hardware support, ElastiStor can scaleup to any extents.

10


How CloudByte ElastiStor is different from legacy storage solutions

The following table explains themajor differences:

Feature Legacy solutions CloudByte ElastiStor

Storage Architecture Monolithic "Patented VSM Architecture"

Tenant Solution Only at access layers At all Storage Stack layers

Full featured tenants No guaranteed QoS, delegatedadministration, or complete secur-ity

Provides guaranteed QoS, delegated administration rights, and com-plete security

Access to tenant-levelresource usage

No Yes

Security Limited Fully secure (complete isolation, access with encrypted key)

Patented VSM Architecture

In a CloudByte Controller, each tenant is completely isolated at storage stack levels and unified as a VirtualStorageMachine (VSM).With complete isolation, VSMs can be treated as if they are allotted dedicated storage controller enabling:

l Shared storage tenants with guaranteedQoS (IOPS, Throughput, and latency), delegated administration rights, and completesecurity.

l Easy shared storagemanagement with access to and control over tenant-level resource usage (disk, cache, network, andCPU).

Together with its rich customizability and scalability, CloudByte ElastiStor offers a powerful storage virtualization solution for ser-vice providers and enterprises.

CloudByte ElastiStor feature highlights

Quality of Service (QoS) within shared storage

CloudByte ElastiStor assures highly predictable performance using its uniqueQoS-centric approach to provisioning. The featurehighlights include:

n Guaranteed throughput, latency, and IOPS: Easily commission storage with guaranteed throughput, latency, andIOPS (Input/Output Operations Per Second) at various levels. You can decide to configure QoS at a Pool level, VSM level, orStorage Volume level.

n Multi-dimensionality: Configure IOPS, throughput, latency, and capacity.

n Grid-like service: Utilize available unused storage resources efficiently.

n On-demand modification: Edit settings at a later point of time without restart.

See Also:"Assuring Quality of Service"

Multi-protocol support

CloudByte ElastiStor provides LUN access using the following protocols:

11


n Fiber Channel (FC)

n iSCSI

n CIFS

n NFS

See Also: "Setting up data access protocols"

Simplified storage provisioning

Assume that you require 20GB storage and 10000 IOPS. With toomany pools and VSMs, manually verifying the apt Pool is a cum-bersome task. Instead, use the auto-provisioning feature in ElastiStor to handle this situation.

ElastiStor provides a simplified storage provisioning where it intelligently provides a list of best-fit pools.

Server disk space is divided into storage spaces based on the number of applications, individualistic requirements, and expectedperformance from each application. With ElastiStor, you can enter the required SLA/QoS parameters and automate the host selec-tion and resource allocation.

See Also: "Provisioning storage"

Delegated administration

Delegated administration lets you determine the actions various administrators take using administrative tools on differentresources.

The administration roles aremapped to functions using security policies. You can create andmaintain separate administration cap-abilities depending on your needs.

See Also: "Delegated administration"

Multi-way High Availability

CloudByte ElastiStor provides N-way High Availability.

See Also: "Configuring High Availability"

Disaster Recovery

CloudByte ElastiStor equips you to effectively recover data in the case of adversities.

Use theMirror feature for remote disaster recovery. When youmirror data, a remote snapshot, a read-only copy of the data on theprimary setup, is created in the secondary storage setup. In the case of disaster, you can activate themirror to take care of the ser-vices handled by the primary setup.

See Also: "Configuring remote disaster recovery"

12


REST APIs for easy integration

CloudByte ElastiStor REST APIs provide a bundle of functions that encompasses all the features in the CloudByte ElastiStorAdministrator Console (CloudByte ElastiCenter). That is, you need not have ElastiCenter to configure andmanage CloudByteElastiStor. Rather you can use your own portal.

The feature highlights include:

n APIs for all the features in ElastiCenter. The APIs function as an end-to-end translation of the Administrative Console.

n APIs are pluggable with delegated administration preset.

n CloudByte provides off-the-shelf plug-ins which you can install and start using.

Intuitive administration console CloudByte ElastiCenter

CloudByte ElastiStor’s centralizedmanagement console CloudByte ElastiCenter makes configuration andmanagement of storagesimple even when you have thousands of applications.

Comprehensively manage your entire storage cluster from a singleWeb-based console. Monitor, control, and audit storage resourceusage down to application-level granularity.

The ElastiCenter features include:

n Wizards that walk you through the entire setup

n Interactive graphical representations that help youmonitor data

13


n Highly usable action pods that let youmanage tasks, with easy access andminimal navigation

See Also:"ElastiCenter User Interface"

Physical storage infrastructure

The following illustration briefs the components of CloudByte ElastiStor:

14


Building your storage infrastructure starts with the creation of a Site, the fundamental infrastructure building block where your Nodeand storage and network infrastructure reside.

The following illustration shows the composition of a Site:

15


See also "Setting up your storage" on page 108

See also " ElastiCenter, ElastiStor's management console" on page 97

Simple hardware setup

The following figure illustrates a simple hardware setup, to form aHA Group, that you can use to get started:

16


Connect the SAS cable between the SAS Port of the Node and Disk Array's INPort . To form a daisy chain, connect SAS cablebetweenOUTPort of one Disk Array and INPort of another Disk array.

This sets up a group of Nodes.

An HA Group can have amaximum of four Nodes.

Installation overview

The installation of CloudByte ElastiStor involves installing

l ElastiCenter. See the procedures in "Installing ElastiCenter"

l Node. See the procedures in "Installing Node"

17


You can install the components separately or together.

Note:Installing ElastiCenter and Node together is not recommended for production environments.

Hardware Compatibility List

As a convenience to our customers and partners, CloudByte comprehensively evaluates a variety of hardware components that canbe used in storage solutions.

The CloudByte Support website summarizes information on hardware components that have undergone complete certification test-ing. As drivers change frequently, please contact CloudByte sales or support teams should you have questions about specific hard-ware or hardware driver versions.

Access the latest Hardware Compatibility List at the following URL: http://www.cloudbyte.com/support/hardware-compatibility-list/

18

http://www.cloudbyte.com/support/hardware-compatibility-list/


Host interoperability matrix

This section is an overview of compatibility of CloudByte ElastiStor with various OS platforms and environments.

Access the host interoperability matrix at the following URL: http://www.cloudbyte.com/support/host-interoperability-matrix/

19

http://www.cloudbyte.com/support/host-interoperability-matrix/

CHAPTER II What's new in ElastiStor 1.4.0 Patch 5


CloudByte ElastiStor 1.4.0 Patch 5 is a stability release includingmajor performance enhancements and various bug fixes. Fol-lowing are the high points of this release:

l Better iSCSI I/O handling

l Usability and supportability improvements

l Performance andQoS improvements

l Improvements in bad disk detection and handling logic

l QoS functionality enhancements for cloned volumes

l Two new monitoring services are added:

l cbc_monitor_pools_ioerror to check the status of Pool I/Os. If the I/Os are suspended, ElastiCenter gen-erates the Alert es.pool.io.suspended.

l CB-Log Monitoring Service to detect disk I/O errors.

What's new in ElastiStor 1.4.0 Patch 4

CloudByte ElastiStor 1.4.0 Patch 4 release includes the following new features:

l "Assigning real IP spaces"

l "Backing up your interfaces"

Assigning real IP spaces

Multiple client virtual machines can access ElastiStor VSMs on same IP addresses. The feature helps in scenarios wheremultiplecustomers need the same network range.

You are not mandated to specify IP address while creating a VSM. Later, you can let the delegated administrator (say, your clientadministrator) specify the IP address. Even if different client administrators specify the same IP address, it will not result in a con-flict.

Scenario

Two customers need the same network range. You can provide the same VSM IP address for both. The feature takes away theneed of IP address conversion.

20


Illustration

Workflow

Assign VLAN to Account

1. Create an Account in ElastiCenter.

2. Add VLAN Interface. For details, see VLAN interface.

3. Assign the VLAN interface to Account (created in step 1).

1. In the Account dashboard, select the Account.

2. In the Actions pane, select Manage Shared NICs.

3. From the list, select the VLAN interfaces that you want to assign to the Account.

4. Click Apply.

Provision VSM

1. Create VSM on the VLAN.

2. When you provision a VSM, in the networking details section of the Provision VSMWizard, specify the VLAN interface youcreated. The list contains VLANs both assigned and unassigned VLANs .

When provisioning a VSM on ElastiCenter, you can choose not to specify an IP address for the VSM.

21

http://www.docs.cloudbyte.com/what-is-elasticenter/provisioning-a-tsm/configuring-interfaces/


When you have provisioned a VSMwithout assigning an IP address, the following page appears:

22


Assign IP address after creating a VSM

1. In the VSMs page, click Advanced Settings.

2. In the Primary IP Address tab, specify an IP address for the VSM.

3. Click Save.

Proceed to provisioning Storage Volumes.

Limitations

l You cannot assign real IP spaces if you have configured VLAN on a LAGG.

l VSMmigration fails if the VSMs are assigned same IP address.

l Theremight be a delay in the deletion of VSMs.

Backing up your interfaces

Back up NIC

For the NIC that you have chosen during the installation, you can set up a backup using ElastiCenter. The redundancy prevents inter-ruption of management network access and connection among the Nodes or between Node and ElastiCenter.

Configuring redundant NIC

Prerequsite

The NIC and the backup NIC should be on the same network.

23


Procedures

1. In the Nodes page, click Network Interface Cards (Actions pane > Network Interface Cards)

2. Click Configure Backup Interface.

3. In the dialog, select a NIC as the backup from the drop-down list.

4. Click OK.

Back up LAGG

For the LAGG that you have added (Add LAGG) you can set up a back up. The following sections explain the procedures. Backing upyour lagg assures that the VSMs are always accessible.

Assumptions

All the NICs in the primary LAGG and backup LAGG should be on the same network.

24



Procedures

1. In the HA Groups page, click Manage LAGG interface (Actions pane > Manage LAGG interfaces).

2. In the Actions pane, click Configure Backup Interface.

3. Specify the following details:

l For LAGG interface 1, select the LAGG for which you want to configure back up.

l In the LAGG interface 2 field, select the LAGG for back up.

4. Click OK.


CloudByte ElastiStor 1.4.0 Patch 3 release includes the following new features:

l "Hardware-based disk encryption/ Support for self-encrypting drives (SED)"

l "Space reclamation"

Hardware-based disk encryption/ Support for self-encrypting drives (SED)

The current release of ElastiStor introduces a hardware-level disk encryption by supporting the self-encrypting drives.

25


l This is a hardware-based data encryption technique

l Encryption workload is moved to the drive instead of the processor. This improves system performance.

l This is a full disk encryption, whichmeans the entire disk data is encrypted (metadata included)

l The data on the disk is encrypted at all times

l The feature enables drive lockingmechanism using the passphrase specified by the Administrator.

l Restricts data access only to the authorized hosts with the specified passphrase.

Enabling Encryption

l Hardware based encryption can be enabled only on SEDs.

l SEDs can be used to create GELI (software) encrypted Pools as well.

l You cannot create an encrypted Pool with a combination of SED and other data disks.

Using ElastiCenter

1. In the Add Pool page, select Enable Encryption.

2. Select Hardware as the Encryption type.

3. Specify a Passphrase.

4. Click Next.

Changing the passphrase

You can change the passphrase used to encrypt Pool by doing the following:

1. Go to the Encrypted Pool page.

2. In the Actions pane, select Change passkey.

3. In the Change Passkey dialog box, specify the new passkey and confirm.

26


Schedule passphrase change


2. In the Actions pane, select Schedule change passkey.

3. In the following dialog box, specify the scheduling interval.

4. Click OK.

Note: CloudByte recommends you to perform aRefresh Hardware after deleting the encrypted Pool.

Space reclamation

On deletion of files on the Client filesystems (ext4/NTFS), you can reclaim space on the thin provisioned iSCSI LUNs. This isachieved by supporting the SCSI UNMAP CDB.

Enabling the space reclamation feature using ElastiCenter

1. In the Storage Volumes page, click the iSCSI Volume for which you want to enable the space reclamation feature.

2. In the Actions pane, click Manage iSCSI configuration.

3. Click Edit.

4. Enable UNMAP by selecting Enable from the drop-down list.

5. Click Save.

Note: CloudByte recommends you to select the option Optimize drives in the disk defragmenter utility onWindows 2012 Cli-ents, to reclaim dead space on the thin-provisioned Volumes.


CloudByte ElastiStor 1.4 Patch 2 release includes the following new features:

27


l "Geli Encryption"

l "NFS home directory"

l "QoS Alerts"

l " GUI changes to distinguish your ElastiStor"

Geli Encryption

In the latest release, CloudByte ElastiStor has introduced a software-level disk encryptionmechanism using the block device-layerdisk encryption system GELI.

The following graphic illustrates the encryption architecture:

l This is a software encryption technique inherited from GELI

l The encryption algorithm used is AES-XTS


l The encryption key used is encrypted using the passphrase specified by the Administrator

l Both the encryption key and passphrase are required to unlock the disks

Enabling Encryption on ElastiCenter



3. Click Next.

28


NFS home directory

The latest release of CloudByte ElastiStor lets you use a NFS share like a home directory. The users on the commonNFS share donot have access to delete the other user files.

To use the NFS Share on ElastiStor like a home directory, perform the procedures in File system NFS options.

Note: Set Map as Root Users to No.

29

http://www.docs.cloudbyte.com/configuring-nfs-protocols/


QoS Alerts

You are alerted when there is a variation in the quality of service (QoS) you have opted for.

Currently, only throttling alerts are available. The alerts are dependent on the following "Global Settings":

l alert.qos.throttling.percentage: Sends an alert if the throttled IOPS percentage exceeds the specified value(default is 20) and if alert.qos.throttle.count is encountered for the specified number of times.

l alert.qos.throttle.count: Sends an alert if the number of consistent throttling exceeds the specified value (default is10) and alert.qos.throttling.percentage exceeds the specified value.

GUI changes to distinguish your ElastiStor

ElastiCenter uses different color schemes for the Community Edition and Enterprise Edition. That difference is visible in log inscreen and the top portion of the page which houses the notificationmenu.

30


31


For details, seeCommunity Edition and Enterprise Edition.


CloudByte ElastiStor 1.4 Patch 1 release includes the following new features:

l "Support for LDAP authentication"

l "Global Settings to provision without specifying IOPS"

l "VSMWebshell terminal on ElastiCenter"

l "Option to check if the Disks areWrite Cache Enabled"

Support for LDAP authentication

You can use an external LDAP server such as Microsoft Active Directory to authenticate ElastiCenter users.To do this, map theElastiCenter administrator types with the groups created in the directory service, for example Active Directory Server groups usingCloudByte ElastiStor global settings.

The following illustration explains how the authentication works:

32

http://www.cloudbyte.com/blog/community-edition-enterprise-edition/


Setting up LDAP authentication involves procedures on both ElastiCenter and the Active Directory Server.

On Active Directory Sever

Create Active Directory user

1. On your Active Directory Server, select Server Manager (Start > Administrative Tools).

2. In the Server Manager, click Tools.

3. In the right pane, click Active Directory Users andGroups.

4. In the Active Directory Users andGroups wizard, right-click your Domain name.

5. Select New and then select User.

33


6. In the User page, specify the relevant information of the user and then click Next.

7. In the following page, specify and confirm password for the user.

8. Click Finish.

Create Active Directory Group

Note: Create an individual group for each administrator type in ElastiCenter. For details, see http://www.-docs.cloudbyte.com/delegated-administration/




4. In the Active Directory Users andGroups wizard, right-click your Domain name.

5. Select New and then select Group.

6. Specify a name for the group to be created.

7. Click OK.

Adding Users to the Groups

Add users to the group that you want to mapwith the users in the various administrator type groups in ElastiCenter.

34

http://www.docs.cloudbyte.com/delegated-administration/





4. In the Active Directory Users andGroups wizard, click your Domain name.

5. From the list of Groups and users, right-click the User you want to add to the Group.

6. Click Add to aGroup.

7. In the dialog box, specify the name of the Group to which you want to add the user.

8. At the prompt, click OK.

On ElastiCenter

1. In the admin section of ElastiCenter, click LDAP Configuration.

35


2. In the Actions pane, click configure LDAP.

36


3. In the Configure LDAP page, specify the configuration details:

Option Description

Host Name The fully qualified domain name or the IP address of the LDAP Server. For example, my.ladap.-com or 10.10.10.10

Port The port on which the LDAP Server is listening. The default port is 389.

Bind DN The distinguished name of the user on the LDAP Server. For example, test\Administrator.

Bind Password Password for the name specified in the Bind DN field.

Search Base Distinguished name of the Directory tree level, where the query should start the search. Forexample, CN=Users,DC=CBQA,DC=com.

4. Click OK.

Set Group names for administrators on ElastiCenter

1. In the admin section of ElastiCenter, click Global Settings.

37


2. In the Global Settings page, search for LDAP.

3. To define the user roles, click Edit and then specify the Group name for the administrators.

4. Repeat step 3 for all types of administrators on ElastiCenter.

Log in to ElastiCenter using the LDAP user credentials

Once you have configured LDAP authentication for CloudByte ElastiStor, youmust add users to the administrator roles inElastiStor. For details of administrator types, see http://www.docs.cloudbyte.com/delegated-administration/.

1. Launch ElastiCenter using the URL in the following format: https://ElastiCenter_Management_IP_Address.

2. Specify the following login credentials of an AD user: username and password.

3. Re-log in to ElastiCenter as super administrator.

38



4. As a super administrator, map the administrator roles to the newly added users.

1. In ElastiCenter, select the Account to which the user belongs to.

2. In the Actions pane, select Manage Administrator.

3. Select Assign Administrator.

Note: If you want to map the administrator roles for Site-Admin or the View-Admin, follow the procedures in the sec-tionCreate Site administrators or view administrators

Global Settings to provision without specifying IOPS

The new global settings option provisioning.uncontrolled.iops lets you enable/disable provisioning with uncontrolledIOPS.

VSMWebshell terminal on ElastiCenter

An option for advanced troubleshooting, where youmight have to access the VSM terminal, for example, to check if your iSCSI ser-vice is running. This option lets youmonitor various processes and services, snapshot list, volume list, etc .

To access the terminal,

1. In ElastiCenter, click Virtual StorageMachines and then select the VSM.

2. In the Actions Pane , click Terminal.

Option to check if the Disks are Write Cache Enabled

The feature lets you see if the disks are write cache enabled.

39



When you Add Pool or Disk group to the Pool, the list of disks displayed shows Yes if disks areWrite Cache Enabled.

What's new in CloudByte ElastiStor 1.4

CloudByte ElastiStor 1.4 release includes the following new/enhanced features:

s

l "Faster and resilient HA"

l "ImprovedQoS Control"

l "8Gbps FC target support"

l "Detecting faulty disks"

l "Replacing faulty disks"

Faster and resilient HA

Significant improvement in HA takeover and giveback time while ensuring data continuity.

Improved QoS Control

Offers tighter control over QoS. Also, you can define throughput for a Storage Volume.

8Gbps FC target support

Now, you can configure 8Gbps FC targets (QLogic).

Detecting faulty disks

When a disk in a Pool goes bad, ElastiCenter displays the bad disk in Alerts as shown in the following screen:

40


The ElastiCenter displays bad disks as seen in the following screen:

l You cannot create Pool on a bad disk.

l You cannot label a bad disk.

Replacing faulty disks

You can now replace a faulty disk from the ElastiCenter.

41


Procedures1. In the Pools page, select a Pool.

2. In the Actions pane, click Manage RAID Configuration.

3. In the Disk Groups page, select a Disk Group in which you want to replace a disk.

4. For the disk you want to replace, click the disk offline icon.

5. At the confirmation prompt click Yes. The red icon in the status indicates the disk is offline.

6. For the offline disk, click the replace icon.

42


7. Select a disk from the drop-down list and then click OK.

Note: A disk can also be replaced without bringing it to offline. But it is recommended tomake the disk offline and thenreplace it.

43

CHAPTER III Installing CloudByte ElastiStor


The following topics provide steps to install CloudByte ElastiStor:

n "Planning your installation"

n "System requirements"

n "Network configuration details"

n "Installation options"

n "Installing ElastiCenter"

n "Installing Node"

The following illustration is a quick-start representation of ElastiStor installation:

Planning your installation

CloudByte ElastiStor installation comprises the following:

n Installing ElastiCenter: Install the administrative infrastructure used tomanage the Nodes.

n Installing Node: Node is the storage infrastructure that is managed by ElastiCenter. You can install only one Node permachine. An HA Group can have up to four Nodes.

Installing on production environments

CloudByte recommends you to install ElastiCenter and Nodes on separately.

Installing on development environments

You can install both ElastiCenter and Nodes together on the samemachine. This is the typical installation used for development andtest setups. Here you play around with test data for evaluation, rather than production where you have business critical data.

44


System requirements

The following topics explain the system requirements for installing CloudByte ElastiStor.

l CloudByte Nnode is built on commodity servers.

n The connectivity between Controller and Storage is either on Ethernet (iSCSI) or on SFF-8088 SAS cables.

n ElastiStor OS is built using FreeBSD as the base platform and therefore CloudByte ElastiStor's Hardware Supported List issimilar to that of FreeBSD Release version.

Minimum hardware requirements to build CloudByte ElastiStorn Supported AMD64, Intel 64 (x86-64), or i386 Processor and System Board (with Dual Core Processor and 8GB of RAM)

n Four 1 Gigabit Ethernet Controller Cards

n One ormore SAS HBAs, depending on the number of disk arrays connected

n OS boot disk drive size of 20 GB. However, the recommended size is 2.4 * RAM size + 33GB

Minimum hardware requirements for CloudByte ElastiCentern Supported AMD64, Intel 64 (x86-64), or i386 Processor and System Board (with Dual Core Processor and 8GB of RAM)

n OS boot disk drive size of 20 GB. However, the recommended size is 2.4 * RAM size + 33GB

n 1Gigabit Ethernet Controller Cards

For supported list of

n AMD64 Processors, see FreeBSD 9.0 supported list of AMD64 processors

n i386 Processors and System Boards of CloudByte ElastiStor, see FreeBSD 9.0 supported list of i386 Processors

n Disk Controllers of CloudByte ElastiStor, see FreeBSD 9.0 supported list of Disk Controllers

n Ethernet Interfaces of CloudByte ElastiStor, see FreeBSD 9.0 supported list of Ethernet Interfaces

Hardware Compatibility List

For the complete list of compatible hardware, seeHardware Compatibility List.

Network configuration details

You need the following network configuration details to proceed. All the options aremandatory.

Option Description

Hostname Name of the host, for example CloudByte.

Root Password The password assigned to the root account.

IP address The IPv4 addresses that you want to access.

Net Mask The IPv4 subnet mask, for example 255.255.255.0.

Gateway The IPv4 gateway IP address, for example 10.0.0.1.

DNS The IPv4 address of the DNS server, for example 8.8.8.8.

Search Domain search, for example lab.engineering.cloudbyte.com.

45

http://www.freebsd.org/releases/9.0R/hardware.html#PROC-AMD64

http://www.freebsd.org/releases/9.0R/hardware.html#PROC-I386

http://www.freebsd.org/releases/9.0R/hardware.html#DISK

http://www.freebsd.org/releases/9.0R/hardware.html#ETHERNET

http://www.cloudbyte.com/support/HCL/


Installation options

The installation wizard provides the following installation options.

n ElastiCenter: Installs only ElastiCenter, which is the central administrative console tomanage your Node. Choose if youprefer to install Node later. For procedures, see "Installing ElastiCenter"

n Node: Installs only the Node, which is managed using ElastiCenter. Choose if you prefer to install ElastiCenter later. For pro-cedures, see "Installing Node"

n Both ElastiCenter and Node: Installs both ElastiCenter and Node. This installation is not recommended for productiondeployments. For procedures, see "Installing ElastiCenter"

Installing ElastiCenter

Note: The steps to install ElastiCenter or ElastiCenter and Node together are similar. The only difference has been detailedin step 4.

The following figure illustrates the ElastiCenter installation workflow:

1. Insert the CloudByte ElastiStor installation disk into your computer's DVD or CD drive and boot up your Node (ensure thatthe disk drive is set as your system's primary boot device). The installation starts up and the following screen with the boot

46


options appear:

By default, the boot option Boot CloudByte ElastiStor is selected. So you need not interfere. To select a different boot option, see"Boot options".

TheWelcome screen appears:

47


2. In the following screen, accept the license agreement.

3. Select the Only ElastiCenter option. For details, see "Installation options".

4. (Applies only if you are installing Both ElastiCenter and Node) Select "Installing ElastiCenter HA".

48


5. (If you havemultiple disks) At prompt, select the disk where you want to install CloudByte ElastiStor and then click OK.

6. (If you havemultiple Ethernet Interfaces) At prompt, select the Ethernet Interface and then click OK.

49


The following screen appears:

7. Specify the "Network configuration details".

8. Click OK. The following screen appears:

9. Select Continue to confirm the settings so that the installation can start up. Tomodify any of the values, select Back. Install-ation starts and the wizard shows the progress.

50


10. In the following screen, specify the region you belong to:

11. Confirm the country and then confirm the time zone.

51


The installation completes and the following screen appears:

After installing the packages, CloudByte ElastiStor reboots (indicated by the following screen).

Wait till the system automatically reboots. The following screen welcomes you after the reboot:

52


The following screen with a login prompt indicates that installation is successful. You can proceed to configure CloudByteElastiStor.

53


Installing ElastiCenter HA1. Follow the procedures in the section "Installing ElastiCenter" till step 3.

2. Select ElastiCenter HA.

3. Do either of the following:

1. Select primary ElastiCenter and then in the following screen, specify the ElastiCenter Management IP address.

54


2. Select Secondary ElastiCenter and then in the following screen, specify the primary ElastiCenter IP address.

4. Click OK.

5. Follow the procedures from step 4 in the section "Installing ElastiCenter" and complete the ElastiCenter installation process.

You can now

1. Launch ElastiCenter in Mozilla Firefox Web browser (by specifying the URL, provided in the "Network configuration details",in the format https://ipaddress).

2. Log in.

3. Perform the configuration procedures. For details, see "Configuring CloudByte ElastiStor".

Boot options

Option Description

Boot CloudByte The default option. Select to start the installation with all standard options enabled.

Escape to loader prompt Lets you perform advanced configuration in a CLI environment.

Reboot Select to reboot.

Boot CloudByte with ACPI Select to boot with advanced configuration and power interface. Helpful for power man-agement.

Boot CloudByte with Safe Mode Diagnostic mode with reduced functionality.

Boot CloudByte with Single User Mode Lets only a single user to login. That is the root user.

Boot CloudByte with Verbose To view behind the scene activities that occur during the installation. Particularly usefulfor troubleshooting .

Installing Node1. Insert the CloudByte ElastiStor installation disk into your computer's DVD or CD drive and boot up your Node Server (ensure

that the disk drive is set as your system's primary boot device). The installation starts up and the following screen with the

55


boot options appear:

By default, the boot option Boot CloudByte ElastiStor is selected. So you need not interfere. To select a different bootoption, see "Installing Node".

TheWelcome screen appears.

2. In the following screen, accept the license information.

56


3. Select the installation option. For details, see "Installation options".

57


4. At prompt, Specify the ElastiCenter IP address and then click OK.

5. (If you havemultiple disks) At prompt, select the disk where you want to install CloudByte ElastiStor.

58


6. (If you havemultiple Ethernet Interfaces) At prompt, select the Ethernet Interface.



8. Click Done. The following screen appears:

59



60


10. In the following screen, specify the region you belong to:


61


The installation completes and the following screen appears:


62



The following screen with a login prompt indicates that installation is successful. You can proceed to configureCloudByte ElastiStor.

Note:When you deploy ElastiStor on production environments, ensure that all disks are SCSI reservation-enabled.

To enable SCSI reservation:

1. Make an SSH connection to the Node.

2. Run the following command: rm /etc/disablescsi.

63


Upgrading to CloudByte ElastiStor 1.4.0 Patch 5

Versions from which you can upgrade

If you are using the following versions of ElastiStor, you can upgrade to ElastiStor 1.4.0 Patch 5.

l 1.4.0.272

l 1.4.0.340

l 1.4.0.528

l 1.4.0.6019

l 1.4.0.7004


You have the following two installation options:

l Upgrade: Upgrades from any of the versions mentioned in "Versions from which you can upgrade" to ElastiStor 1.4.0 Patch5.

l New Installation: Replaces the current version of ElastiStor with ElastiStor 1.4.0 Patch 5.

When you upgrade, ensure that you upgrade ElastiCenter first, and then proceed to upgrade the Node.

The workflow is as follows:

1. Back up ElastiCenter. For details see the section "Backing up configurations".

2. Upgrade ElastiCenter using the steps in the following section Upgrade procedures.

3. Log in to the Node as root and then run the following command to download the Volume configuration script: fetchhttp://cloudbyte.com/downloads/updateVolSize.py

4. Run the following command: python updateVolSize.py

5. Move the Node toMaintenanceMode. For details see the section Bringing Node tomaintenancemode in "Configuring HighAvailability".

6. Upgrade the Node using the procedures in the following section Upgrade procedures.

7. Move the Node to Available mode. For details, see the sectionMaking a Node available in "Configuring High Availability".

8. Refresh hardware for the upgraded Node.

9. Ensure that the Pools on the Node are online.

10. Repeat the procedures (from step 3 to step 7) for the other Node.

Upgrade procedures1. Boot up your server and insert the CloudByte ElastiStor installation disk into your computer's DVD or CD drive. The install-

ation starts up and the following screen with the boot options appear:

64


By default, the boot option Boot CloudByte ElastiStor is selected. Do not interfere.


65



3. Select Upgrade in the following screen:

The upgrade progresses as shown in the following screen:.

66


The upgrade completes and the following screen appears:


67



68


The following screen with a login prompt indicates that upgrade is successful. You can proceed to configureCloudByte ElastiStor.

69


You can now


2. Log in.


release Moving your CloudByte ElastiStor setup to a new network environment

Assume that you have tomove your existing ElastiStor setup to a different location or network. For example, you have created asetup on staging environment and have performed all the tests. Now your setup is ready for production.

In this case, you canmove your Storage setup to the production site using the procedures explained in this section.

70


Procedures1. Move all Nodes toMaintenancemode. For details, see "Configuring High Availability"

2. On ElastiCenter server, change the network configurations of ElastiCenter. For details, see "releaseMoving your CloudByteElastiStor setup to a new network environment ""releaseMoving your CloudByte ElastiStor setup to a new network envir-onment "

3. Change the network configurations of Node. For details, see "Change the network configuration details of Node"

71


4. In the ElastiCenter database, update the network configurations of Nodes. For details, see "releaseMoving your CloudByteElastiStor setup to a new network environment "

5. In ElastiCenter, change the IP address of the HA group. For details, see "Creating an HA Group ".

6. Restart the Nodes.

7. Bring all the Nodes back to Available mode. For details, see "Configuring High Availability"

8. (Optional) Change the network configuration setup for VSM. For details, see "Provisioning a VSM".

Change the network configurations of ElastiCenter

1. Log in to ElastiCenter as root.

2. In the /usr/local/cb/bin/, run the following command: python Change_network_config.py

3. Perform the steps as shown in the following sample command line :

Choose an option1)Change the network configuration details of ElastiCenter2)Update the network configuration details of Node

1

Here, you have to select 1 as highlighted

You are going to change some or all of the network configuration details that you specified at thetime of installation.

Specify the New IP address for ElastiCenter20.10.83.57To change the Netmask, specify the details in the dotted decimal format. To bypass this step, specifyNo255.0.0.0To change the Gateway, specify the details. To bypass this step, specify No20.10.83.1To change the DNS, specify the details. To bypass this step, specify No8.8.8.8To change the Network Interface card, specify the details. To bypass this step, specify Noem0To change the Domain name, specify the details. To bypass this step, specify Nocloudbyte.com

The new network details of ElastiCenter are as follows:IP address: 20.10.60.50Netmask : 255.0.0.0Gateway: 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain Name: cloudbyte.com

Review the details and confirm by specifying Yes. Specify No to exit and cancel the changes. Tochange the details later, rerun the Change_network_config script.YesNetwork details have been changed. Update the same in the associated Nodes.Reboot ElastiCenter

4. Reboot ElastiCenter using the following command: reboot

Note: Unless you update the ElastiCenter's new IP address on the Nodes, ElastiCenter displays the Node status as offline(though the Node is up and accessible).

Change the network configuration details of Node

1. Log in to Node as root.


72




To change the IP address, specify the details. To bypass this step, specify No20.10.83.57To change the Netmask, specify the details in the dotted decimal format. To bypass this step, specifyNo255.0.0.0To change the Gateway, specify the details. To bypass this step, specify No20.10.83.1To change the DNS, specify the details. To bypass this step, specify No8.8.8.8To change the Network Interface card, specify the details. To bypass this step, specify Noem0To change the Domain name, specify the details. To bypass this step, specify Nocloudbyte.comTo update the Elasticenter IP address, specify the details. To bypass this step, enter No

Here, you have to specify the new IP address of ElastiCenter if changed,else specify No.

The new network details of ElastiCenter are as follows:IP address: 20.10.83.57Netmask : 255.0.0.0Gateway: 20.10.83.1DNS: 8.8.8.8Network Interface Card: em0Domain Name: cloudbyte.com

Review the details and confirm by specifying Yes. Specify No to exit and cancel the changes. Tochange the details later, rerun the Change_network_config script.

Yes

Network details have been changed.Reboot the Node

4. Reboot the Node using the following command: reboot

Update the network configurations of Node in the associated ElastiCenter

1. Log in to ElastiCenter as root.



Choose an option1)Change the network configuration details of ElastiCenter2)Update the network configuration details of Node

2

Here, you have to select 2 as highlighted

You are going to update ElastiCenter with the network configuration details of one of the followingNodes:

+-------+--------------+| Node | Old IP address|+-------+--------------+| Node1 | 20.10.83.60 |+-------+--------------+

73


| Node2 | 20.10.83.6 |+-------+--------------+

Specify the old IP address of the Node20.10.83.60Specify the New IP address of the Node20.10.83.57To change the Netmask, specify the details in the dotted decimal format. To bypass this step, specifyNo255.0.0.0To change the Gateway, specify the details. To bypass this step, specify No20.10.83.1To change the DNS, specify the details. To bypass this step, specify No8.8.8.8To change the Network Interface card, specify the details. To bypass this step, specify Noem0To change the Domain name, specify the details. To bypass this step, specify Nocloudbyte.com

The new network details of ElastiCenter are as follows:Old IP address: 20.10.83.60New IP address: 20.10.60.57Netmask : 255.0.0.0Gateway: 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain Name: cloudbyte.com

Review the details and confirm by specifying Yes. Specify No to exit. To update the details later,rerun the Change_network_config script.

Yes

Network details have been updated in ElastiCenter.Reboot ElastiCenter

4. Reboot ElastiCenter using the following command: reboot

After completing the network configurationmodifications, go back to the section "Procedures" and complete the procedures.

Tips

Ensure the following:

l A physical connectivity between the Nodes and ElastiCenter exists using the command ping.

l The IP address and Subnet are configured correctly using the command ifconfig -a .

l TheGateway is configured correctly using the command netstat -rn.

74

CHAPTER IV CloudByte ElastiStor Appliance


This chapter applies to only the ElastiStor Appliance users.Skip this chapter if you are

l Using only the CloudByte ElastiStor OS

l Evaluating ElastiStor with the trial version downloaded from the CloudByte website

The following topics help you get started with CloudByte ElastiStor Appliance:

l "Installing ElastiCenter and ElastiStor"

l "Moving your CloudByte ElastiStor Appliance (ESA) to a new network environment "

Installing ElastiCenter and ElastiStor

The following figure illustrates the CloudByte ElastiStor installation workflow for ESA:

Prerequisitesl The following details are required before you start the installation:

IP address for Description

Management The IP address used for accessing the ElastiCenter.

Primary ElastiCenter Host IP address of the Node where primary ElastiCenter has to be installed.

Secondary ElastiCenter Host IP address of the Node where secondary ElastiCenter has to be installed.

75


Note:ElastiCenter Management IP address and ElastiCenter Host IP address cannot be the same.

Installing primary ElastiCenter1. Insert the CloudByte ElastiStor installation disk into your computer's DVD or CD drive and boot up your Node.

Note: Ensure that the disk drive is set as your system's primary boot device.

The installation starts up and the following screen with the boot options appears:

By default, the boot option Boot CloudByte ElastiStor is selected. So you need not interfere. To select a different boot option,see "Installing ElastiCenter and ElastiStor".

TheWelcome screen appears:

76


2. In the following screen, accept the license agreement.

77


3. Select the option Both ElastiCenter and Node.


78


4. Select ElastiCenter HA. The following screen appears:

5. Select primary ElastiCenter and then in the following screen, specify the ElastiCenter Management IP address (IP addressused to access the ElastiCenter).

6. Click OK.

79


7. (If you havemultiple disks) At prompt, select the disk where you want to install CloudByte ElastiStor and then click OK.

8. (If you havemultiple Ethernet Interfaces) At prompt, select the Ethernet Interface and then click OK.

80




10. Click OK. The following screen appears:


81


12.1. In the following screen, specify the region you belong to:


82


14. The installation completes and the following screen appears:

Installing secondary ElastiCenter1. Perform the steps 1 to 4 in the section "Installing primary ElastiCenter".

2. In the following screen, select secondary ElastiCenter.

3. Specify the primary ElastiCenter's Host IP address.

83


4. Perform the steps 6 to 14 in the section "Installing primary ElastiCenter" to complete the secondary ElastiCenter installationprocedures.

Note:After the installation, ensure that the primary ElastiCenter is active prior to working with the Secondary ElastiCenter. Ifsecondary ElastiCenter is active before the primary ElastiCenter, you will have to reboot secondary ElastiCenter.

Moving your CloudByte ElastiStor Appliance (ESA) to a new network environment

When you get started with ElastiStor Appliance, the initial step is to configure the network settings. You have tomodify some or allof the default network configuration settings depending on your environment.

All CloudByte ElastiStor Appliancemodels have the following default network configurations. Go through the following default net-work settings so that you can change the required values.

For example, if your scenario might require only changing the IP addresses. In that case, you have to replace only the values10.11.12.3 (Management IP address), 10.11.12.1 (Primary Node), and 10.11.12.2 (Secondary Node). You can leave all other valuesintact.

Network detail Description Default

Management IP address The IP address used for accessing the ElastiCenter. 10.11.12.3

Primary Node IP address Host IP address of the Node where primary ElastiCenter is installed. 10.11.12.1

Secondary Node IP address Host IP address of the Node where secondary ElastiCenter isinstalled.

10.11.12.2

Netmask The standard IPv4 subnet mask. 255.0.0.0

Gateway The IPv4 gateway IP address. 10.11.12.254

DNS The standard IPv4 address of the DNS Server. 8.8.8.8

84


Domain name Domain search, for example lab.engineering.cloudbyte.com. cloudbyte.com

Health check IP range Health check IP range of the HA Group. 10.11.12.11-10.11.12.12

IPMI address of primary Node You can remotely log in to primary Node. 10.11.12.101

IPMI address of secondaryNode

You can remotely log in to secondary Node. 10.11.12.102

Default log in credentials:

Component User name Password

ElastiCenter admin password

IPMI ADMIN ADMIN

SSH root test

Procedures1. Bring the Nodes in the appliance (ESA-A50 and ESA-A100) to maintenancemode. For details, see "Configuring High Avail-

ability"

2. (On the secondary Node) In the /usr/local/cb/bin/, run the following command: python Change_network_con-fig.py


You have logged into the secondary Node and the associated ElastiCenter.


Specify the new IP address for the secondary Node20.10.60.50

To change the Netmask, specify the details in the dotted decimal format. To bypass this step, specifyNo. No

To change the Gateway, specify the details. To bypass this step, specify No. No

To change the DNS, specify the details. To bypass this step, specify No. No

To change the Network Interface card, specify the details. To bypass this step, specify No. No

To change the Domain name, specify the details. To bypass this step, specify No. NoSpecify the management IP address used to access ElastiCenter20.10.39.71Specify the primary Node IP address20.10.60.40

The new network details of the secondary Node are as follows:

IP address: 20.10.60.50Netmask : 255.0.0.0Gateway: 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain Name: cloudbyte.comManagement IP address: 20.10.39.71Primary node IP address:20.10.60.40


85


Network configuration details changed successfully.Reboot the Node.

4. Reboot the secondary Node using the following command:reboot.

5. (On the primary Node) In the /usr/local/cb/bin/, run the following command: python Change_network_con-fig.py

6. Perform the steps as shown in the following command line :

You have logged into the primary Node and the associated ElastiCenter.


Specify the new IP address for the secondary Node20.10.60.50

To change the Netmask, specify the details in the dotted decimal format. To bypass this step, specifyNo. No

To change the Gateway, specify the details. To bypass this step, specify No. NoTo change the DNS, specify the details. To bypass this step, specify No. NoTo change the Network Interface card, specify the details. To bypass this step, specify No. NoTo change the Domain name, specify the details. To bypass this step, specify No. NoSpecify the management IP address used to access ElastiCenter20.10.39.71

The new network details of the primary Node are as follows:

IP address: 20.10.60.50Netmask : 255.0.0.0Gateway: 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain Name: cloudbyte.comManagement IP address: 20.10.39.71


Specify the following network configuration details of the secondary Node:IP address: 20.10.60.50Netmask: 255.0.0.0Gateway : 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain name: cloudbyte.com

The network details of the secondary Node are as follows:IP address: 20.10.60.50Netmask: 255.0.0.0Gateway: 20.10.1.1DNS: 8.8.8.8Network Interface Card: re0Domain Name: cloudbyte.comIP address used to access the Management Console: 20.10.39.71Review the details and confirm by specifying Yes.Specify No to exit and cancel the changes.To change the details later, rerun the Change_network_config script.Network configuration details changed successfully.

Reboot the Node.

7. Reboot the primary Node using the following command:reboot.

86


8. After the system is rebooted,launch ElastiCenter using the following steps:

1. Launch aWeb browser and specify URL in the following format:

https://New IP Address to access the management console

2. Specify the following default credentials:

Credentials Value

User name admin

Password password

9. In ElastiCenter, change the IP address of the HA group. For details, see "Creating an HA Group "

10. Bring the secondary Node to Available mode. for details, see "Configuring High Availability"

11. Bring the primary Node to Available mode. for details, see "Configuring High Availability"

12. In the ElastiCenter admin pane, select ElastiCenter HA.

87


Ensure that both the Nodes are in sync with each other as shown in the following screen:

Upgrading to CloudByte ElastiStor 1.4.0 Patch 5

Versions from which you can upgrade

If you are using the following versions of ElastiStor, you can upgrade to ElastiStor 1.4.0 Patch 5.

l 1.4.0.272

l 1.4.0.340

l 1.4.0.528

l 1.4.0.6019

l 1.4.0.7004


You have the following two installation options:

88


l Upgrade: Upgrades from any of the versions mentioned in "Versions from which you can upgrade" to ElastiStor 1.4.0 Patch5.

l New Installation: Replaces the current version of ElastiStor with ElastiStor 1.4.0 Patch 5.

Prerequisitesl Either have IPMI access or physical access to ESA.

l Both the Nodes in ESA are in sync with each other (see step 12 in the following sectionWorkflow).

l Back up ElastiCenter and Node configurations. For details, see Back up configurations.

Assumptions

The document assumes the following:

l Node 1 is the primary Node and hosts Pool 1.

l Node 2 is the secondary Node and hosts Pool 2.

Workflow Node 1(Primary) Node 2(Secondary)

1 a. Bring the Node to Maintenance mode

b. Power off the Node.

2 Becomes Primary, all configurations from Node 1 aretaken over.

3 a. Ensure that the configurations are taken over

b. Insert the product CD and boot as DVD.

4 Perform the procedures in "Upgrade procedures". On suc-cessful upgrade, the Node comes online as secondary Node .

5 Move the Node to Available Mode.

6 a. Bring the Node to Maintenance mode

b. Power off the Node.

7 Becomes Primary, all configurations from Node 2 are takenover

8 a. Ensure that the configurations are taken over

b. Insert the product CD and boot as DVD.

9 Perform the procedures in "Upgrade procedures".On successful upgrade, the Node comes online as sec-

ondary Node

10 Move the Node to Available Mode.

11 In the ElastiCenter admin pane, select ElastiCenter HA.

12 Ensure that both the Nodes are in sync with each other as shown in the following screen (screen 1).

13 Confirm that the version number (provided in the release notes) matches the one provided in the following screen(screen 2).

89

http://www.docs.cloudbyte.com/backing-up-elasticenter-configurations/


Screen 1:

Screen 2:

Upgrade procedures1. Boot up your server and insert the CloudByte ElastiStor installation disk into your computer's DVD or CD drive. The install-

ation starts up and the following screen with the boot options appear:

90


By default, the boot option Boot CloudByte ElastiStor is selected. Do not interfere.


91



3. Select Upgrade in the following screen:

The upgrade progresses as shown in the following screen:.

92


The upgrade completes and the following screen appears:


93



94


The following screen with a login prompt indicates that upgrade is successful. You can proceed to configureCloudByte ElastiStor.

95


You can now


2. Log in.


96

CHAPTER V ElastiCenter, ElastiStor's management console

CHAPTER V ElastiCenter, ElastiStor's management con-sole

Introduction

ElastiCenter is the centralizedmanagement tool that you use to configure, monitor, manage, and deploy the services provided byCloudByte ElastiStor.

ElastiCenter lets you

n Use theGraphical User Interface tomanage the storage environment

n Generate statistical and configuration reports to help troubleshoot

n Delegate administration tasks

n Track events

n Globally control various settings

Using ElastiCenter, the entire ElastiStor deployment, consisting of multiple geographical sites spanning clusters and servers, canbemanaged from a single browser session. Storagemanagement tasks such as High Availability and disaster recovery can be donefrom the same session.

To illustrate, let us examine the processing of a file recovery request for a particular storage tenant using a legacy solution andCloudByte ElastiStor:

Legacy solution CloudByte ElastiStor

The administrator has to

1. Manually recover the file from the DR stor-age system

2. Manually transfer to the primary site

3. Merge with the primary site.

In CloudByte ElastiStor, using ElastiCenter, the administrators can performactions remotely from the management portal. The administrator can, fromthe same browser session,

1. Log in to the primary site.

2. Identify the storage file system that needs to be recovered.

3. Browse for the corresponding backup copies on the remote DR site.

4. Recover data with a single click.

The following illustration shows the workflow to setup a storage environment using CloudByte ElastiCenter:

97


ElastiCenter User Interface

The following topics orient you with the ElastiCenter User Interface.

Navigation pane

The navigation pane in the left-hand side of ElastiCenter provides a list of options.

When you select an option, you are taken to a specific component page. Use the component pages to perform various configurationandmanagement tasks.

98


Note:The options displayed in the navigation pane depend on the administrative privileges you have. If you are a superadministrator, you have all the options. For further details, see "Managing CloudByte ElastiStor".

Content pane

The content pane displays the component pages that provide options and data to configure CloudByte ElastiStor. For example, ifyou select Sites in the navigation pane, the Sites page appears.

99


Summary pane

Summary pane sums up the number of storage components, such as Sites or Pools, in the storage infrastructure.

Actions pane

Action pane, in the right-hand side of the content pane, provides a list of configuration options. For example, in the HA Groups page,go the Actions pane and add Node.

100


Notification menu

The informationmenu in the top right-hand side of ElastiCenter lets you access Alerts, Events, Reports, and Administrator data.You also have the option to go back to the Home page.

See Also:"Monitoring" and "CloudByte ElastiStor reporting"

Admin options

The Admin options help you with various administrative tasks.

101


See Also:CloudByte ElastiStor API Reference, "Licensing", and "Delegated administration"

Launching CloudByte ElastiCenter

CloudByte ElastiCenter is a browser based console which you can launch using one of the popular browsers such as Mozilla Firefoxor Google Chrome.

Note:Ensure that you have updated Adobe Flash Player to themost recent version.

To access ElastiCenter,

1. LaunchMozilla Firefox Web browser.

2. Specify the URL in the following format: https://IP Address

Note:The IP Address you specify is the one that you configured in the installation screen. See "Installing ElastiCenter" fordetails.

Authentication

You need credentials (username and password) to log in to CloudByte ElastiCenter.

1. After installation is completed, launch ElastiCenter using the URL in the following format: https://IP Address.

102

http://www.elastistor.info/


2. Specify the following default credentials :

Credentials Value

Username admin

Password password

After you log in, you can change the password.

Reset password and configure auto-alert options

After you log in, you are prompted to:

l Change your password

l Update your profile. CloudByte Technical Support might use this information to contact you when your storage infrastructureencounters any issues.

l Review the auto-alert details that CloudByte Technical Support collects automatically. Technical Support might use theinformation to resolve potential problems that your storage infrastructure encounters. You have the option not to share par-ticular information.

To reset password and configure auto-alert options,

103


1. Login to ElastiCenter using the default credentials (admin|password). The following screen appears:

2. Specify a new password and then confirm.

3. Update your profile.

4. Review the auto-alert information, and uncheck the options that you do not want to share.

5. Click Update.

Replacing default SSL Certificate

The connection between ElastiCenter and browser is secured using SSL. If required, you can replace the default certificate usingthe following procedures.

Note:The utility used to perform the following tasks is keytool. You can use a utility of your preference.

104


1. In the ElastiCenter console, run the command as shown in the following example: keytool -genkey -aliasprimarycb -keyalg RSA -keystore mysprimary.keystore -keysize 2048. At prompt, provide passwordand the domain details.

2. In the ElastiCenter console, generate Certificate Signing Request (CSR) by executing the command as shown in the fol-lowing example:keytool -certreq -keyalg RSA -alias primarycb -file mycertreq.csr -keystoremyprimary.keystore. You are then prompted to specify a password for keystore.

3. Get the CSR signed by the trusted CA. (for example, VeriSign or Thawte).

4. (Optional) Append the certificate with intermediate CA and trusted root CA.

5. Import the CA's certificate into your keystore as a trusted certificate using the following command:keytool -importcert -file verisign.cer -keystore myprimary.keystore -alias primarycb

6. Copy the keystore to the directory /usr/local/apache-tomcat-6.0.33/conf.

7. Edit the following lines in the file server.xml in the directory /usr/local/apache-tomcat-6.0.33/conf/

keystoreFile="conf/myprimary.keystore"keystorePass="cloudbyte"

8. Run the following command to shut down Tomcat:/usr/local/apache-tomcat-6.0.33/bin/shutdown.sh

9. Kill Java processes using the following command: pkill java

10. Execute the following command to ensure that no Java process is running:pgrep java

11. Start Tomcat using the command /usr/local/apache-tomcat-6.0.33/bin/startup.sh

Customizing the CloudByte ElastiCenter theme

This section will help you customize CloudByte ElastiCenter with minimal modifications in CSS files.

Edit the theme.css file to make customizations1. The company logo above the navigation pane (left pane) can be changed by replacing the image in the folder images/cloud-

byte. Name your image as company-logo. To avoid distortion in the image, ensure that the image does not exceed the fol-lowing dimensions: 215 px width, 67 px length.

2. The company logo on the left corner of the notificationmenu can be changed by replacing the image in the folderimages/cloudbyte. Name your image as edit-logo. Ensure that your image is in PNG format. To avoid distortion in theimage, also ensure that the image does not exceed the following dimensions: 166 px width, 24 px length.

3. To change the header border in ElastiCenter, edit the highlighted line in the following section:

4.

#dashboard-header_wrap{

background: #E7E8E7;

position:relative;

z-index:3;

border-top:4px, solid #ed1652;

}To change icons on the notificationmenu, replace the following images (referred to in the screenshot) in theimages folder.Ensure that all your images are in PNG format.

Image to be replaced Name the image as

A home

B alert

105


C events

D reports

5. To change the active tab top border, edit the highlighted line in the following section:

.nav ul li:hover,.active-nav{

background:#fff;

border-top:4px solid #bcdf5a!important;

}

6. To change the widget top border, edit the highlighted line in the following section:

.tcl-graph{

margin:0 12px 15px 3px;

border-top:4px solid #BCDF5A;

}

106


Modify the index.jsp file to include theme.css file

For the CSS changes to take effect, include the theme.css file in index.jsp file located in the following path: /us-r/local/apache-tomcat-6.0.33/webapps/client/.

Add the highlighted section as shown in the following sample snippet:

<head>

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />

<title>CloudByte ElastiCenter 1.3</title>

<link type="text/css" rel="stylesheet" href="css/theme.css" />

<link type="text/css" rel="stylesheet" href="css/<filename>.css" />

<script src="lib/jquery.js" type="text/javascript">

</script>

<script src="lib/jquery-ui/js/jquery-ui.js" type="text/javascript">

</script>

</head>

107

CHAPTER VI Configuring CloudByte ElastiStor


CloudByte ElastiStor is configured andmanaged using the administrator console namedCloudByte ElastiCenter. The following top-ics explain the configuration andmanagement tasks that you can perform.

n "Setting up your storage"

n "Creating a Site "

n "Creating an HA Group "

n "Adding a Node"

n "Configuring Disk Array enclosure"

n "Creating a Pool"

n "Creating an Account"

n "Setting up data access protocols"

n "Provisioning a VSM"

n "Configuring VSMmigration"

n "Provisioning storage"

n "Choosing a Pool to provision VSM"

n "Modifying QoS after provisioning VSM"

n "Configuring High Availability"

n "Configuring local backup"

n "Configuring remote disaster recovery"

n "Assuring Quality of Service"

Setting up your storage

In CloudByte ElastiStor, the set of configurations that you perform can be divided in to tasks related to

n Infrastructure, the base structure of your storage service

n Accounts, that you create for storage specific needs

The following illustration shows how your infrastructure and account are organized in CloudByte ElastiStor:

108


The following sections explain the concepts in detail:

Site

Building your storage infrastructure starts with the creation of a Site, the fundamental infrastructure unit where your Node and stor-age and network infrastructure reside.

The following figure depicts the structure of a Site:

The following figure explains each component of the Site:

109


Accounts

After you set up the storage infrastructure, build the account infrastructure. That is, you set up the client/customer for using your stor-age infrastructure.

The following figure explains how an account is constituted in CloudByte ElastiStor:

110


The following figure explains each component associated with an account:

111


Creating a Site

Use the Sites page (ElastiCenter > Sites) to create a Site.

The Sites page provides a graphical representation of the Sites that you have created. It provides the Site details such as HAGroups, Pools, Provisioned Storage, and VSMs that you have set up.

Click any of the Site details link for specific review and configuration. For example, click HA Groups to access details of the HAGroups associated with the Site.

To create a Site1. In the Sites page, click Add Site in the actions pane.

The Add Site page appears.


Field Description

Name Unique name for the site.

Location Where the site is located.

Description A meaningful description for the Site.

3. Click Add.

Creating an HA Group

Use the HA Groups page (ElastiCenter > HA Groups) to create an HA Group.

All Nodes in the HA group you createmust be on the same

112


l Public network: used for management of Nodes.

l Private network: used with VSMs for data transfer.

Creation of HA Groups leads to the formation of High availability Clusters. For details, see "Configuring High Availability"

1. In the HA Groups page, click Add HA Group.

The Add HA Group page appears.

2. Specify the following options and then click Add.

Field Description

Site The Site to which you add the HA Group.

Name Unique name for the HA Group.

Description Meaningful description for the HA Group.

IP Address Range Provide a range of dedicated IP addresses in the management network. For instance, ensurethat you do not specify the IP address used for Node, ElastiCenter, VSM, or client. Provide thefirst and last IP address in the range, for example 1.1.1.100 and 1.1.1.104. The IP addressesyou specify are used to check the heartbeat between nodes.

Changing CARP IP address range1. Bring Node1 tomaintenancemode.

2. Bring Node2 tomaintenancemode.

3. In the HA Groups page, select the HA Group for which you want to change the CARP IP range.

4. Click Settings in the actions pane.

113


5. In the following page, click edit and then change the details.

6. Click Save.

Adding a Node1. In the ElastiCenter navigation pane, click Nodes.

2. In the actions pane, Click Add Node.

3. In the Add Node page, specify the following details:

Field Description

Site Select a Site.

114


HA Group Select the HA Group to which the Node has to be added.

Name Unique name for the Node.

IP Address IP address of the Node.

4. Click Add Node.

Deleting a Node from an HA Group

Follow the procedures to delete a Node from a single Node HA Group:

1. Delete all the Storage Volumes on the Node.

2. Delete all the VSMs on the Node.

3. Delete all the Pools on the Node.

4. Bring down the Node tomaintenancemode.

5. Click the delete icon to remove the Node from HA Group.

Follow the procedures to delete a Node from anHA Group with more than one Node:

1. Bring down the Node tomaintenancemode.

2. Click the delete icon to remove the node from HA Group.

Configuring Disk Array enclosure

After you add Node, you are redirected to the Node page.

When there is no Disk Array attached to the Node, the following page appears. Ensure that you attach a Disk Array to the Node andthen click Refresh Hardware.

Note:When you attach a new Disk Array to a Node, click Refresh Storage/Refresh Hardware tomake the Disk Array visibleon the ElastiCenter.

115


To configure a newly attached Disk Array enclosure, follow the procedures:

1. In the Node page, click the image in the section Shared Storage.


116


2. Click Configure A New Disk Array. The Disk Array Configuration page appears:

3. Specify the Name, Bay Alignment, Number of columns and Number of Rows. For details, see the "Configuration details".

4. Click Next and then do the following:

l Click the option under the column Blink. For details, see "SCSI Enclosure Service ".

l Specify the bay location for the disks from the drop down list under the column Bay.

117


5. Click Apply. On successful configuration, the Node page appears with a representation of the shared storage:

Note:You can also partially configure a Disk Array Enclosure. Suppose you have 12 disks, you can initially configure around8 disks and other 4 disks can be configured as per the requirement.

Configuration detailsHorizontal and vertical alignments

The Disk Arrays can be idealized either in terms of horizontal or vertical alignments.

The following figure represents a vertical alignment:

118


The following figure represents a horizontal alignment:

Slots per row

Slots per row is the number of disks each row can accommodate. For example, three slots per row implies that a single row canaccommodate three disks.

SCSI Enclosure Service

SCSI Enclosure Service (SES) helps monitor the status of each device in a disk array enclosure such as disk slots, fans, heatsinks, or power supplies.

The service helps you

119


l Identify a disk or verify the slot in which the disk is located.

l Locate and disable a faulty disk.

l With specific and general alerts on enclosure components.

Blink identification

Blink identification LED on the disk slot of a disk array. When you click Blink, at the physical location of the disks, the identificationLED in the specified disk slot blinks. If the disk is not in the SCSI enclosure, the LED blink does not occur.

If the disk is in the SCSI enclosure and still if the blink does not occur, verify the health of the disk.

Enclosure component alerts

CloudByte ElastiStor provides common and specific alerts for the enclosure components such as disk slots, power supplies,voltage sensors, current sensors, temperature sensors, fans, and audible alarm.

The common alerts which apply to all components include the ones related to

l Removal and installation of the component

l State in which the component is, such as critical, non-critical, OK, or unrecoverable

Specific alerts are generated for the following enclosure components: power supply, voltage sensor, current sensor, and tem-perature sensor.

By default, CloudByte sends alerts over mail to the email you specified when you updated your profile the first time you logged in toCloudByte ElastiCenter. To expand themailing list, add the IDs separated by comma in the Global settings (Information Pane >Admin > Global Settings > alert.email.addresses).

To view alerts in CloudByte ElastiCenter, click Alerts in the information pane.

Deleting a Disk Array1. Delete all Pools in the Disk Array.

2. In the HA Groups page, select the HA group from which Disk Array has to be deleted.

3. For the Disk Array to be deleted, click Manage Disk Array and then click the delete icon.

Creating a Pool1. Go to the Node page (ElastiCenter > Nodes).

2. From the list of Nodes, select the one for which you create the storage Pool.

3. In the actions pane, go to the section Tasks and then click Add Pool.

120


The Add Pool page appears.

4. Select the Site, HA Group, and CB Node to which the Pool belongs.

5. Specify a name for the Pool and then click Next.

6. In the following screen, specify the RAID storage type and then choose one or more of the available disks.

7. (Optional) Enable Sector Size.

Note: For optimal performance, ensure that the disks in the same Pool belong to the same disk array.

121


8. Click Next. TheQoS settings page appears.


Field Description

122


Total IOPS The maximum per second I/O limit for a Pool.

AverageLatency

The desired average latency in milliseconds per I/O. Latency is the round-trip time per I/O request-response as measuredby the client.

10. Click Add Pool.

Note:The number of disks required to create a Pool depends on the RAID type and protection/performance you look for. Forexample, for a high performance application, you can create RAID 10 Pool. Similarly, for a high capacity application, you canset up RAID z1 with 8 disks.

Configuring Spare1. Go to the Pools page (Navigation pane > Pools).

2. In the Actions pane, click Manage RAID Configuration. The Disk Group page appears.

3. In the Actions pane, select Add Spare.

4. In the Add Spare page, select a disk from the drop-down list.

5. Click Add.

Migrating a Pool

You cannot migrate a pool. Only a VSM can bemigrated.

Pool encryption

GELI encryption

This is a software-level disk encryptionmechanism using the block device-layer disk encryption system GELI.

The following graphic illustrates the encryption architecture:

123


l This is a software encryption technique inherited from GELI

l The encryption algorithm used is AES-XTS


l The encryption key used is encrypted using the passphrase specified by the Administrator

l Both the encryption key and passphrase are required to unlock the disks

Enabling GELI encryption on ElastiCenter



3. Click Next.

Hardware-based disk encryption/ Support for self-encrypting drives (SED)

This is a hardware-level disk encryptionmechanism using the self-encrypting drives.

l This is a hardware-based data encryption technique

l Encryption workload is moved to the drive instead of the processor. This improves system performance.


l The data on the disk is encrypted at all times

l The feature enables drive lockingmechanism using the passphrase specified by the Administrator.

l Restricts data access only to the authorized hosts with the specified passphrase.

SED-based encryption conditions

l Hardware based encryption can be enabled only on SEDs.

l SEDs can be used to create GELI (software) encrypted Pools as well.

124


l You cannot create an encrypted Pool with a combination of SED and other data disks.

Enabling SED-based encryption on Elasticenter


2. Select Hardware as the Encryption type.


4. Click Next.

Changing the passphrase

You can change the passphrase used to encrypt Pool by doing the following:


2. In the Actions pane, select Change passkey.

3. In the Change Passkey dialog box, specify the new passkey and confirm.

Schedule passphrase change1. Go to the Encrypted Pool page.

2. In the Actions pane, select Schedule change passkey.

3. In the following dialog box, specify the scheduling interval.

4. Click OK.

Note: CloudByte recommends you to perform aRefresh Hardware after deleting the encrypted Pool.

Creating an Account

Use the Account page (ElastiCenter > Accounts) to create an Account.

125


1. Click Add Account. The Add Account page appears.


Field Description

Name Unique name for the account.

Description Meaningful description of the account.

Assign Admin Select the type of administrator that you want to create. When you select an administratortype, the username and password fields appear. For details, see "Delegated administration".

3. Click Add Account. The Account is created. The wizard prompts you to continue and provision VSM.

4. Click Provision VSM to continue.

Delete account user1. In ElastiCenter, select Accounts in the navigation pane.

2. From the list of Accounts, select the Account from which you want to delete a user.

3. Go to the Tasks section of the actions pane and then select Account Users.

4. Click the delete icon for the account user that you want to delete.

5. At prompt, click Yes.

Provisioning a VSM

Use either the Create Account wizard or Virtual StorageMachines page (ElastiCenter > Virtual StorageMachines) to provisionVSM.

1. Click Provision VSM.

126


The Provision VSM page appears.


Field Description

Account The Account for which you create the VSM.

Name A unique name for the VSM.

Capacity The amount of disk space to be allotted for the VSM.

127


Total IOPS The maximum per second I/O limit for the VSM.

TotalThrough-put

When you specify the Total IOPS, CloudByte calculates the Total Throughput and is automatically added. You cannot editthis field.

3. Click Next and then specify the provisioning details by selecting a Pool from the list.

4. Click Next.

5. Select the network interface details. For details, see "VLAN interface".

6. Specify the tenant IP address and default gateway.

128


7. Click Provision.

Multipathing

Multipathing is the ability to configure a server with multiple paths to the storage for take over on network failure. CloudByteElastiStor supports multipathing for iSCSI. When network failure occurs, redundant network takes over and ensures that the iSCSILUN is available.

Note: SATA disks or any disk with SATA interface is not supported for HA andmultipathing configuration.

The details of the supportedMaximumMultiple Connections (MCS) per LUN are as follows:

n 8 connections per LUN

n 16 sessions per LUN

To enablemultipathing,

1. In ElastiCenter, select Virtual StorageMachine.

2. In the VSM page, select the VSM for which you want to enablemultipathing.

3. In the action pane, select Advanced Settings.

4. Click Edit.

129


5. Specify the required details.

130


Field Description

Secondary IP address Provide secondary IP address to enable multipathing.

Secondary Interface Select an interface from the list.

6. If partial failover of Pool is enabled (which is enabled by default), click yes at the prompt to disable it.

Configuring VSMmigration

The objective of VSMmigration is to optimize QoS. VSMmigration helps to control the QoS parameters within a Pool and tomovedata.

You canmigrate a VSM

l From a Pool to another within the sameNode or to a different Node.

l Across Sites.

Note: When youmigrate a VSM across Nodes, ensure that the Node to which themigrating VSM belongs to is not down.

1. In the VSM page, select a VSM from the list.

2. In the Actions pane, go toMigrate and then click Schedule.

3. In the following page, specify the required details:

131


Field Description

Site Select a Site to migrate the VSM.

IOPS The IOPS of the primary VSM is populated by default. You can modifythe value to provision IOPS required for migration.

Throughput The throughput of the primary VSM is populated by default. You cannotmodify this value as it is depends on the values of IOPS and Block SIze.

4. Click Next.

5. Select the Pool to migrate VSM and then click Next.

132


6. Select the Backup interface, IP Address, and schedule of migration (in the format minutes/hours/days/month/year).

7. Click Add to createmigrant VSM.

133


Migrant VSM activation

Activation considerationl TheMigrant VSM can be activated only after the base snapshot is successfully transferred from the primary VSM. You can

check the status of data transfer by clicking View Transfer Status (Actions > Tasks).

l The backup interface and IP address range of the native and themigrant VSMs must be same.

Note: After activation, this backup IP address will not be active and therefore need to be removed at the Pool level.

1. In themigrant VSM page, click Settings in the Actions pane.

2. In the Settings page, click Edit and then select an interface from the drop-down list.

3. Click Save.

134


4. Click the VSM, theMigrant VSM, and then click Activate.

5. At prompt, click Yes.

After the VSMmigration is active, the re-mount prompt appears. Ensure that you re-mount the Storage Volumes before you proceed.

VLAN interface

VLAN interface can be created on a NIC whenever there is a need for the interface to subscribe to a separate broadcast domain (usu-ally needed for high traffic data networks).

Additionally, multiple VLAN interfaces can be created on a single physical NIC, thereby allowing access tomultiple networks at thesame time. The VLANs act as virtual individual NICs, each belonging to a separate broadcast domain.

135


Adding VLAN interface

1. In the HA Groups page (ElastiCenter > HA Groups), go to the actions pane and then click Tasks > Manage VLAN Interfaces.

2. In the VLAN Interfaces page, go to the actions pane and then click Tasks > Add VLAN Interface.

3. In the Add VLAN Interface dialog box, specify the VLAN tag (whichmust be a numeric value) and then select the parent NIC.

4. Click OK.

Defining the VLAN Interface for the VSM

When you provision a VSM, in the networking details section of the Provision VSM Wizard, specify the VLAN interface you created(in the Interface field).

136


LAGG interface

LAGG interface allows aggregation of multiple network interfaces into a single virtual LAGG interface, providing fault-tolerance andhigh-speedmulti-link throughput. The aggregation protocols supported by LAGG determines which ports are used for outgoing trafficand whether a specific port accepts incoming traffic.

Adding LAGG interface

1. In the HA Groups page (ElastiCenter > HA Groups), go to the actions pane and then click Tasks > Manage LAGG Interfaces.

2. In the LAGG Interfaces page, go to the actions pane and then click Tasks > Add LAGG Interface.

3. In the Add LAGG Interface dialog box, specify the following details:

l LAGG ID: Unique id (numeric value) for the LAGG interface.

l Protocol Type:

l Failover: Select Failover to send traffic only through the active port. If themaster port becomes unavailable,the next active port is used. The first interface added is themaster port. Any interfaces added after that areused as failover devices. By default, received traffic is only accepted when received through the active port.

l Load Balance: Balances outgoing traffic across the active ports based on hashed protocol header inform-ation and accepts incoming traffic from any active port. This is a static setup and does not negotiate aggreg-ation with the peer or exchange frames tomonitor the link

l LACP:LACP negotiates a set of aggregable links with the peer in to one or more Link Aggregated Groups.Each LAGG is composed of ports of the same speed, set to full-duplex operation. The traffic will be balancedacross the ports in the LAGGwith the greatest total speed, in most cases there will only be one LAGGwhichcontains all ports.

137


4. Click Next.

5. From the list of available NICs, drag and drop the ones you want to assign to a LAGG.

6. Click Add.

Provisioning storagel Before you provision a Storage Volume, youmust create a VSM.

l Themaximum supported size of the Storage Volume depends on the available capacity of VSM.

l There is no limit to the number of Storage Volumes that you can create. But the composite capacity of the created StorageVolumes must be less than or equal to the capacity of the VSM they belong to. For example, if the total capacity of the VSMis 1 TB, then you can create a Storage Volumewith the capacity of 1 TB ormultiple storage volumes where the total capacityof all volumes is less than or equal to 1 TB.

l You can enable both compression and deduplication on the same Storage Volume.

l File system storage (NFS and CIFS) works together on the same Storage Volume, but block level storage (iSCSI) does not.

l You can back up Storage Volume by creating snapshots and then rolling back when required. For details, see "Configuringlocal backup".

Note: By default, ElastiStor Storage Volume is thin provisioned. You can expand the Storage Volume up to the size of thePool, provided the Pool has available storage.

Use the Storage Volumes page (ElastiCenter > Storage Volumes) to provision storage.

138


1. Click Provision Storage.

2. The Add Storage Volume page appears.


Field Description

Account Account for which you provision the storage.

Name Unique name for the storage.

Capacity Amount of disk space to be allocated.

139


Provision Mode Shared or Dedicated. For details, see the section Difference between Shared and Dedicated in "QoS FAQs".

Block Size Block size of the volume.

AverageLatency

Desired average latency in milliseconds per I/O. Latency is the round trip time per I/O request-response.

Control IOPS Select the required IOPS using the slider. Qos can be enforced either throughonly IOPS or only Throughput.

ControlThroughput

Select the required throughput using the slider. Qos can be enforced either throughonly IOPS or only Throughput.

4. Click Next and then select the VSM onwhich you want to provision the Storage Volume.

140


5. Click Next. The screen to specify the advanced settings appears.


Field Description

Volume Hierarchy The VSM on which the Storage Volume is mounted.

Record Size The block size to which data has to be divided.

Synchronization Leave the default settings as it is.

Deduplication If ON, new data is deduplicated.

Compression If ON, new data is compressed.

Mount Point (Applies only for NFS and CIFS) Specify the Storage Volume mount point for the client to use the storage.

Enable NFS Select to access the Storage Volume using NFS protocol.

Enable CIFS Select to access the Storage Volume using CIFS protocol.

Enable iSCSI Select to access the Storage Volume using iSCSI protocol.

7. Click Provision.

Note:Disconnect all associated clients before you delete an iSCSI Storage Volume.

141


Choosing a Pool to provision VSM

CloudByte ElastiStor follows aQOS-centric approach to provisioning. When you provision a VSM, your choice of the Pool dependson the availability of Pools that satisfy the required throughput and IOPS.

Selecting the Pool1. In the ElastiCenter navigation pane, click Virtual StorageMachine.

2. Click Provision VSM.

3. After you specify the VSM details such as Name and theQoS details, click Next.

4. In the following screen, select the Pool after evaluating the IOPS.

5. Click Next to specify the Network details.

Choosing a VSM to provision Storage Volume

When you provision storage, your choice depends on the availability of VSMs that satisfy the required throughput and IOPS.

To provision storage,

1. In ElastiCenter, click Storage Volumes in the navigation pane.

2. Click Provision Storage.

3. In the following screen, specify the basic storage details and then click Next.

142


4. Select a VSM that satisfies the QoS requirements. The graphical representation helps you with the IOPS data.

5. Click Next and then proceed to complete the provisioning.

Auto-Provisioning

ElastiStor can automate Node selection and resource allocation if you provide the required SLA/QoS parameters. Sensing the levelof storage performance and capacity the application requires, ElastiStor automatically allots the right Pool.

Auto-provisioning is particularly helpful when you have innumerable Pools/VSMs. CloudByte ElastiStor chooses the appropriatePool/VSM for provisioning automatically.

How to auto-provision

1. In ElastiCenter, select Global Settings.

2. Click the Edit icon for provisioning.sb.auto and provisioning.tsm.auto.

3. In the Value column, specify yes (instead of the default nowhich implies provisioning is manual) as shown in the followingscreen:

143


Example

Assume that auto-provisioning is enabled and there are three Pools (pool1with 500 IOPS, pool2with 400 IOPS, and pool3with300 IOPS respectively). If the administrator wants to create a VSMwith 100 IOPS, pool1(the Pool with the highest IOPS value) isautomatically selected for VSM creation.

SAS multipathing

CloudByte ElastiStor supports SAS multipathing. Havingmultiple physical paths to the disk array enclosure assures effective pathfailover by providing redundant paths for data access.

l By default, SAS multipathing is enabled.

l SAS multipathing is an active/passive setup

l At any point in time, there is only one active path for I/O

l If a failure occurs, instead of the active path, the one that was redundant (so far) takes over avoiding any interruption to dataaccess

144


Prerequisites

l Two LSI SAS616X switches

l LSI SAS2008 HBA (at the Node side)

l Firmware: 07.15.08.00

l Driver: 15.00.00.00

l SAS cables

l Disk array enclosure with multiple IN ports (for example Dell MD1220)

l Controllers (for example Dell R720)

l In the Node page, clear all existing pools to avoid Pool name conflicts. To do this, first configure the Disk Array in the Nodepage. Then select the Node and then at the bottom of the page, click Clear for the Pools to be deleted.

Sample setup

The following figure shows a simple SAS multipath setup:

145


Verifying the setup in ElastiCenter

When you have themultipath set up, the paths are visible in the related Node page.

1. In CloudByte ElastiCenter, click Nodes.

2. In the Nodes page, select the Node and then click the Disk Array in Shared Storage. The Disk Table provides the details ashighlighted in the following figure:

In the case of a failure, as shown in the following instance the path disappears from the table:

146


The figure implies that there is only a single SAS path available.

Modifying QoS after provisioning VSM

Assume that you have provisioned storage for a VSM and now you want to change theQoS settings. For example, your pool has anIOPS limit of 100. You have provisioned VSMwith 50 IOPS. You canmodify the values.

Note: In the example, you can increase the value up to 100 (whichmeans, the entire IOPS is consumed).

Tomodify the QoS settings,

1. In ElastiCenter, select Virtual StorageMachines.

2. Select the VSM for which you want to modify the QoS settings.

3. In the QoS settings section of the actions pane, click Basic Settings. The highlighted section displays the QoS-related set-tings.

147


4. Click the Edit icon and thenmodify the editable fields.

5. Click Apply.

Modifying QoS after provisioning Storage Volume1. In ElastiCenter, select Storage Volumes.

2. Select the Storage Volume for which you want to modify the QoS settings.

3. In the Storage Volume page, click Settings.

148


4. Click the edit icon and thenmodify the QoS details.

5. Click Save.

Modifying storage capacity1. In ElastiCenter, select Storage Volumes.

2. Select the Storage Volume for which you want to modify the storage capacity.

149


3. In the following page, click Settings.

4. In the following page, click Edit and thenmodify the quota.

5. Click Save.

Note: The quota of the Storage Volume cannot exceed the VSM size.

Configuring High Availability

CloudByte ElastiStor supports High Availability in two-way, three-way, and four-way configurations.

Note: SATA disks or any disk with SATA interface is not supported for HA andmultipathing configuration.

Before you proceed

Ensure the following on all Nodes in the HA Group:

l SameNIC (for example, if em0 is the interface used on one Node, the same should apply for all Nodes) and same capacityDisks are shared (across the Nodes)

l Same system hardware configuration

l Identical SAS HBA (identical firmware and driver)

l Identical FC HBA (identical firmware and driver)

Procedures

The following procedures explain how to configure a two-way High Availability setup:

1. Set up your hardware. For details see "Hardware setup diagram for High Availability"

2. Create a Site. For details, see "Creating a Site ".

3. Create an HA Group. For details, see "Creating an HA Group ".

When you specify the IP address range while adding the HA Group, ensure that you specify a range that matches the numberof nodes you add.

4. Add a Node (Node1). For details, see "Adding a Node".

5. Label shared disks. For details, see "Configuring Disk Array enclosure"

Note: You can perform this step later, after you add all other Nodes. In that case, you have to refresh the hardware(Nodes > select specific Node > select Refresh Hardware in the Tasks section of the action pane) for labeling to takeeffect.

6. Add Node 2. For details, see "Adding a Node".

Now you have established a two-way HA. The following screen provides a visual representation of the configuration.

150


Proceed to create Pools, VSMs, and Storage Volumes.

151


Hardware setup diagram for High Availability

Bringing Node to maintenance mode

When you want to service your Node, change its status tomaintenancemode. In this mode, the Node is up and controlled by theadministrator, but is not an active part of the High Availability Group.

To bring a Node tomaintenancemode,

152


1. Select the Node that you want to bring tomaintenancemode (ElastiCenter > Nodes).

2. In the Node Page, go to the actions pane and then click Move tomaintenance icon.

The Change State of Node dialog box appears.

3. Select Maintenance as the state.

4. Optionally specify a reason for changing the state and then click OK.

Take Over

WhenNode failure occurs, takeover happens as illustrated in the following screens:

n The following screen depicts the Node Page that lists a two-way HA. Here, Node2 is in maintenancemode.

153


n In the following screen, Node1 has taken over.

Note: In ElastiStor, every Node has a designated partner. If the Node goes down, the resources of that Node are failed overto the partner Node.

Making a Node available

After a downtime, when you bring up the Node, you have tomanually change the status to Available.

1. Select the Node that you want to bring up (ElastiCenter > Node).

2. In the Node page, go to the actions pane and then click Move to available icon.

154


3. Select Available as the state.

4. Optionally specify a reason for changing the state and then click OK.

HA Jobs

To view the details of the past High Availability events, perform the following procedures;

1. (In ElastiCenter)In the HA Groups page, select an HA Group for which you want to view event details.

2. In the Actions pane, click View Jobs.

You can view the details of the past HA events and the time taken to perform these events as shown in the following screen:

155


Configuring local backup

Backup refers to copying and archiving of data. In the event of data loss, you can use the backup to restore the original data.

You can create a

n Snapshot, which is a read-only virtual copy of your Storage Volume and VSM.

n Clone, which is a real copy of your Storage Volume

In CloudByte ElastiStor, backup operates as follows:

n The first backup is a full backup.

n The first backup is followed by incremental backups based on snapshot technology.

ElastiStor supports continuous production during snapshot recovery of a Storage Volume.

Backing up VSM1. In ElastiCenter, select Virtual StorageMachines in the navigation pane.

2. From the list, select the VSM to create local backup.

3. In the following page, go to the Local Backup section of the Actions pane and then select either of the following options:

n Schedule: Schedule your backup by choosing a frequency.

n Snapshot:Manually create a snapshot on demand.

Add snapshot

1. In the Local Backup section, click Create Snapshot.

2. In the Create Snapshot dialog box, specify a unique name.

3. Click OK.

156


Add schedule

1. In the Local Backup section, click Add Snapshot Schedule. The Add Snapshot Schedule page appears.

2. Specify the following details and then click Add.

Field Description

Name A unique name for the backup schedule.

Retention Copies Specify the number of backup copies that you want to create.

Schedule Select the frequency.

Backup best practice

As a best practice, CloudByte recommends amultiple backup schedule.

For example, you can configure amultiple backup schedule comprising

Frequency Retention copies

Every 15 minutes 4

Every hour 24

Every day 30

Manage snapshots and schedules

To view the snapshots and schedules that you have created, click Manage Snapshots/Manage Snapshot Schedules in the localbackup section. You have options to rollback and disable the schedules/snapshots.

Backing up Storage Volume1. In ElastiCenter, select Storage Volume in the Navigation pane.

2. From the list, select the Storage Volume to create snapshot.

3. In the following page, go to the Local Backup section of the actions pane and then click Create Snapshot.

157


4. In the Create Snapshot dialog box, specify a unique name for the snapshot and then click OK.

Restoring Snapshots1. In ElastiCenter, select Storage Volumes in the Navigation pane.

2. From the list, select the Storage Volume to restore data.

3. In the following page, go to the Local Backup section of the actions pane and then click Manage Snapshots.

4. From the list of snapshots, select the one that you want to restore, click the restore icon, and then confirm at prompt.

158


Add schedule

1. In the Local Backup section, click Add Snapshot Schedule. The Add Snapshot Schedule page appears.

2. Specify the following details and then click Add.

Field Description

Name A unique name for the backup schedule.

Retention Copies Specify the number of backup copies that you want to create.

Schedule Select the frequency.

Manage snapshots and schedules

To view the snapshots and schedules that you have created, click Manage Snapshots/Manage Snapshot Schedules in the localbackup section. You have options to rollback and disable the schedules/snapshots.

Cloning the Storage Volume

You can create a READ/WRITE copy of your Storage Volume using the clone option. Unlike Snapshot which is a read-only virtualcopy, clone is a real copy of your Storage Volume. It consumes the same disk space as the parent Storage Volume.

ElastiStor uses snapshot as the point of reference for your clone. The workflow is as follows:

1. Create a Storage Volume.

2. Create a snapshot.

3. Create a Clone of the Storage Volume using Snapshot as the point of reference.

Creating a clone

1. In ElastiCenter, select Storage Volume in the navigation pane.

2. From the list, select the Storage Volume to create a clone.

3. In the following page, go to the actions pane and then select Local Backup > Create Snapshot.

4. In the Create Snapshot dialog box, specify a name and then click OK.

159


5. Click Manage Snapshot (actions pane > Local Backup).

6. From the list of snapshots, go to the one which you want to use as reference for creating the snapshot and then click theClone icon (highlighted in the circle).

7. Specify a unique clone name, where tomount the clone, and then click OK.

Configuring remote disaster recovery

The disaster recovery mechanism in CloudByte ElastiStor helps you effectively retrieve data in disaster situations.

Currently, ElastiStor supports disaster recovery for a single Site. You can rebuild a primary Site using the disaster recovery copy.ElastiStor provides disaster recovery across Sites overWANs.

Setting up remote disaster recovery1. In the ElastiCenter navigation pane, click Virtual StorageMachines.

2. From the list of VSMs, select the one for which remote disaster recovery has to be configured.

3. In the actions pane, click Create DR VSM in Remote Backup.

Note: You have this option only if you have not created a DR VSM. If a DR VSM is created, you canmodify the settings.

160


4. In the Add DR VSM page,

1. Select the Site on which you want to configure the DR VSM.

2. Specify a name for the DR VSM.

3. Click Next.

5. In the following page, select a Pool and then click Next.

6. Select the Backup interface, IP Address, and specify a schedule for the DR VSM (in the format minutes/hours/days/-month/year).

7. Click Add.

A message appears to confirm that the DR VSM has been added.

Managing remote disaster recovery

For a specific VSM, use the Remote Backup section in the actions pane to configure remote disaster recovery.

1. In the ElastiCenter navigation pane, click Virtual StorageMachines.

2. From the list, select the VSM tomanage remote disaster recovery.

3. In the actions pane, use the options in the section Remote Backup.

161


Disaster recovery options

n Name: Provides details of DR VSM and displays the transfer details.

n Disable Transfer: Select and confirm to disable disaster recovery.

n Edit Schedule: Edit the disaster recovery schedule.

n Delete DR VSM: Remove a disaster recovery setup that you have configured.

Recovering data

Note: The DR VSM can be activated only after the base snapshot is successfully transferred from the primary VSM. Youcan check the status of data transfer by clicking View Transfer Status (Actions > Tasks).

1. In the ElastiCenter navigation pane, click Virtual StorageMachines.

2. From the list of VSMs, select the DR VSM that is associated with the Primary VSM where the disaster occurred.

3. In the DR VSM page, go to actions pane and then click Tasks > Activate.

162


4. In the following page, specify the network settings of the DR VSM and then click OK.

5. At confirmation prompt, click Yes.

After you activate the DR VSM, the re-mount prompt appears. Ensure that you re-mount the Storage Volumes before you proceed.For details, see "Connecting to ElastiStor".

Upgrading a High Availability DR setup

Follow the procedures to seamlessly upgrade a High Availability disaster recovery environment. Node 1 and Node 2 imply the nodeson primary cluster. Similarly, DR Node 1 and DR Node 2 imply the nodes on the secondary cluster:

1. Upgrade ElastiCenter. For details, see "Upgrading to CloudByte ElastiStor 1.4.0 Patch 5"

2. Move the Node 1 tomaintenancemode. For details, see "Configuring High Availability".

3. Upgrade Node 1. For details, see "Upgrading to CloudByte ElastiStor 1.4.0 Patch 5"

4. Bring the Node 1 to Available mode. For details, see "Configuring High Availability".

5. Move the Node 2 toMaintenancemode.

6. Upgrade Node 2.

7. Bring the Node 2 to Available mode.

8. Move the DR Node 1 toMaintenancemode.

9. Upgrade DR Node 1.

10. Bring the DR Node 1 to Available mode.

11. Move the DR Node 2 toMaintenancemode.

163


12. Upgrade DR Node 2.

13. Bring the DR Node 2 to Available mode.

Confirm if the DR Transfer occurs after the upgrade as scheduled. In the case of any exceptions, contact CloudByte Technical Sup-port.

Assuring Quality of Service

Introduction

Assuring quality of servicemakes your storage performance predictable. CloudByte ElastiStor is unique in its QoS offering as it letsyou

n Easily commission storage with guaranteed IOPS at various levels. You can decide to configure QoS at a Pool level, VSMlevel, or Storage Volume level.

n Configuremultidimensional QoS involving IOPS, throughput, latency, and capacity.

n Utilize untapped system resources. For example, assume that you have three VSMs but only one is active. In this case, theactive VSM can use the entire system’s capability.

n On-demandmodification of settings without system restarts.

Choosing your QoS

The following points help you choose an appropriate QoS:

Endpoint: Decide at what level you want to implement QoS.

Select a capable pool: Based on VSM’s requirements, you provision it on a capable Pool that canmeet the requirements.Define the pool level QoS, based on Pool configuration and the controller level. QoS is determined based on the Pool capacity,which is the underlying storage capability.

QoS provisioning: Understand your VSM’s QoS requirements across the data access and data protection needs. Based on therequirement of the VSM, you can either chooseQoS automatically or manually. Ensure that you are aware of the Pool settings andhow effectively you want to use it at the VSM level.

Sub-divide QoS: Within a VSM, based on varying needs, further sub-divide QoS across Storage Volume, based on the guaranteethat the administrator wants to set. For example, when a VSM is used to store logs in the database, since logs need higher IOPSand lower throughput and latency, set it accordingly. Whereas a database requires low IOPS, high throughput, and high latency.

Configuration

Use ElastiCenter to configure QoS settings. The configuration can be at the following levels:

n Pool: For details, see "Creating a Pool".

n Account: For details, see "Creating an Account".

n VSM: For details, see "Provisioning a VSM".

164


Configuring Grace

What is Grace

Grace is the provisioning of unused IOPS/throughput of Pools to Storage Volumes based on the performance requirements. You canconfigure Grace at Pool level and Storage Volume level.

For example, assume that you have a Pool with the capacity of 100 IOPS andGrace enabled. Two VSMs named VSM A and VSMB, with a capacity of 30 IOPS and 60 IOPS respectively are created under the Pool.

l Under VSM A (with 30 IOPS capacity), Storage Volumes named E and F are created, and assigned 10 IOPS and 15 IOPSrespectively.

l Under VSM B (with 60 IOPS capacity), Storage VolumeG is created and assigned 35 IOPS.

l The grace available in this scenario is 40 IOPS.

Configuring Grace at the Pool level

Grace is enabled by default at the Pool level. You can edit Grace after creating the Pool.

To edit,

1. Go to the Pool page and then in the actions pane, click Settings.

2. Click Edit and then choose to enable/disable Grace.

165


Configuring Grace at the Storage Volume level

To enable Grace at the Storage Volume level, ensure that Grace is enabled at the respective Pool level.

To edit,

1. Go to the Storage Volume page and then in the actions pane, click Settings.

2. Click Edit and then choose to enable/disable Grace.

166


3. Click Save.

Monitoring QoS

In the Storage Volume page, select the Volume for which you want to monitor QoS. The page that appears displays themonitoringdata.

QoS tips

Storage capacity

Discs, storage connectivity, and type

Capability: Ensure that you are well aware of the capabilities of your storage subsystem. For example, the right interconnectionbetween the controller and storage, the type of disks on the disk array, and the ability to expand the capacity.

Limitations: Ensure that you have assessed the limitations. For example, increasing storage capacity does not make any senseif the network does not support.

167


Controller characteristics

RAM, CPU, and network:

CPU: If the storage is more I/O bound than CPU bound, it is more effective. CPU load should not bemore than 60-70 percent, oryou addmore cores.

RAM: RAM determines the ability to cache. Larger RAMmeans more filesystem caches. In a READ-centric scenario, larger RAMhelps generate better QOS. That is, more I/O and throughput but lesser latency.

Network: Network capacity decides data access and data protection. Assure adequate network bandwidth to export performanceto clients. Network should do justice to the storage system.

Segregation: Segregate data access and data protection so that protection workloads do not interfere with data access work-loads.

Pool creation

Storage group tips

As a best practice, create storage with 8 to 10 disks.

When you add storage groups, keep them identical. For example, do not add disparate groups to the same Pool. For example, if onegroup has SATA disks and another group has SAS disks, then do not assign both to the same Pool. That is, do not create hybridPools.

Multiple Pools within a Controller

Tips to avoid contention points: Be judicious when you addmultiple Pools to a controller. For example, it does not makesense if network bandwidth is already utilized to themaximum with the existing Pools, or CPUs are already busy serving the exist-ing Pools. QoS also suffers because of over provisioning.

When to combine multiple pools:When no conflict pertaining to bandwidth, CPU, or Memory exist, you can combine Pools.

Optimizing QoS

Disk characteristics: Better the disk characteristics, better the QOS. For example, the IOPS, throughput, and latency of SATAdisks are not as good as that of SS disks.

Disk layout: Better the disk layout, better the QoS. For example, keep the disk layout to 8 to 10 disks.

Improving WRITE performance: If the application requires synchronous WRITES, associate high performing disk to thePool to speed up theWRITEs. This guarantees better QOS.

Improving READ performance: To improve the READ performance or increase the filesystem cache hit rate, add high per-forming disks as cache devices to the Pool.

168

CHAPTER VII Setting up data access protocols


CloudByte ElastiStor provides LUN access using the following protocols:

l Fiber Channel

l iSCSI

l CIFS

l NFS

When you create a Storage Volume, you are provided with an option to select the access protocols that you want to enable.

Deleting a Storage Volume disables the protocol access services automatically.

The following sections help you configure data access protocols supported by CloudByte ElastiStor:

n "Data access using iSCSI"

n "Data access using CIFS"

n "Data access using FC "

n "Data access using NFS"

Data access using iSCSI

iSCSI is Internet SCSI (Small Computer System Interface), an Internet Protocol (IP)-based storage networking standard for linkingdata storage connections. By offering SCSI commands over IP networks, iSCSI is used to transmit data over intranets. For furtherdetails, see http://en.wikipedia.org/wiki/ISCSI.

iSCSI concepts

Authentication Group: A grouping of multiple iSCSI users.

Use this option if you use CHAP ormutual CHAP. You can configure Authentication group at both VSM and Storage Volume levels.

CHAP: An authenticationmethod which uses a user/secret or IQNs for outgoing authentication. The Storage Volumes onCloudByte ElastiStor authenticate the initiator.

Mutual CHAP: A super set of CHAP which uses peer user/peer secret for outgoing/incoming authentication between StorageVolumes and initiator.

Initiator group: Defines who can initiate the LUN.

It is a combination of multiple IQNs. By default, CloudByte ElastiStor creates two initiator groups (all where everyone can initiate theLUN and none where no one can). After you create the Storage Volume, when youmanage the iSCSI options, you are prompted tochoose an initiator group. You can either select the default one or the ones that you create. You configure initiator group in the Stor-age Volume.

169

http://en.wikipedia.org/wiki/ISCSI


iSCSI configuration workflow

You configure iSCSI at the following levels:

n Account

n VSM

n Storage Volume

The following figure illustrates the work flow for configuring the iSCSI protocol options:

170


Account iSCSI options

Configuring iSCSI at account level involves the following tasks:

1. In ElastiCenter, select Accounts.

2. In the Accounts page, select the account for which you want to specify the protocol options.

3. In the action pane, select the protocol option and then specify the details.

171


The following options are available:

Manage iSCSI Authentication Groups

In the Add Authentication Group dialog box, specify the following details:

Field Description

Name A unique name for the authentication group.

Description A meaningful description for the authentication group.

CHAP Username Username for CHAP authentication. For uniqueness, specify the initiator IQN name as username.

CHAP Secret Password for CHAP authentication in alpha-numeric format. The password must have 12 to 16 characters.

Mutual CHAP Username Username for mutual CHAP authentication. For uniqueness, specify the initiator IQN name as username. Thevalue must be same as that of the CHAP user.

Mutual CHAP Secret Password for mutual CHAP authentication (should be different from CHAP secret). The password must have12 to 16 characters.

Initiator Groups

In the Add Intiator Group dialog box, specify the following details:

Field Description

Name Unique name of the initiator group.

Allowed Intiator List of allowed initiators, separated by comma.

Authorized Networks List of allowed networks followed by subnet mask in the format 10.0.0.0/8, sep-arated by comma.

VSM iSCSI Options

This is the set of iSCSI options specific to a VSM. By default, CloudByte ElastiStor creates a default VSM iSCSI options groupdefaultwhich contains industry standard options.

After you create the VSM, when youmanage the iSCSI options, you are prompted to choose the VSM iSCSI options group. You caneither select the default one or the ones that you create.

Field Description

IQN Name Specify a unique IQN.

Number ofworker threads

Select the number of worker threads.

Discovery AuthMethod

Specify the authentication method.

Discovery AuthGroup

Specify the authentication group.

Timeout Sets the limit on how long an I/O can be outstanding before an error occurs. Values range from 0 to 300. Thedefault value is 30.

NOPIN Interval The frequency of sending a NOP-IN packet in seconds. The values range from 0 to 300. The default value is 20.

172


MAX R2T Specify a value between 1 and 255. The default value is 32.

Max Sessions Specify a value between 1 and 64. The default value is 8.

Max Con-nections

Specify a value between 1 and 64. The default value is 16.

First BurstLength

Maximum data in bytes that an iSCSI initiator may send to the target during the execution of a single iSCSI commandunsolicited. Values range from 1 to 2^32 with a default of 65,536.

Max BurstLength

MaximumWRITE size in bytes the target is willing to receive between R2Ts. Values range from 1-2^32. The defaultvalue is 262,144.

Max ReceiveData Segmentlength

In bytes. Values range from 1 to 2^32. The default value is 262,144.

Max Out-standing R2T

Maximum number of ready to receive packets (R2Ts) the target can have as outstanding. This is for a single iSCSIcommand,where larger values should yield performance increases until MaxOutstandingR2T exceeds the size of thelargest Write I/O divided by MaxBurstLength. Values range from 1 to 255. The default value is 16.

Default time towait

Minimum time in seconds to wait before attempting a logout or an active task reassignment after an unexpectedconnection termination/reset. Values range from 1 to 300. The default value is 2.

Default time toretain

Maximum time in seconds after Time2Wait before which an active task reassignment is still possible after an unex-pected connection termination/reset. Values range from 1 to 300. The default value is 60.

Volume iSCSI Options

This is the set of iSCSI options specific to a Storage Volume. By default, CloudByte ElastiStor creates a default Volume iSCSIoptions group default which contains industry standard options. After you create the VSM, when youmanage the iSCSI options,you are prompted to choose the Volume iSCSI options group. You can either select the default one or the ones that you create.

Field Description

Status Select either Enabled or Disabled.

IQN Name Specify the unique IQN.

AuthenticationMethod

Specify the authentication method.

AuthenticationGroup

Specify the authentication group.

Initiator Group Specify the initiator group.

Initial Digest Select one of the advanced data integrity options.

iSCSI Block Length Select the block length.

Queue Depth Pending I/O requests for the Volume. The values range from 0-255. 0 implies that the status disabled. Thedefault value is 32.

Configuration scenarios

The following scenarios explain iSCSI configuration in detail:

173


Scenario 1: Assigning the same iSCSI initiator groups to two Storage Volumes under a VSM

iSCSI initiator groups can be configured for only Storage Volumes based on initiator groups in the account. For example, StorageVolumes under the VSM can have different iSCSI initiator groups.

For procedures, see "Configuring initiator group and restricting access"

Scenario 2: Assigning different authentication groups to multiple Storage Volumes under a VSM

You can have different authentication groups for VSMs and Storage Volumes.

For procedures, see "iSCSI authentication workflow"

iSCSI access control workflow

An initiator group defines who can initiate the LUN. After you create the Storage Volume, when youmanage the iSCSI options, youare prompted to choose an initiator group.

174


Data access to the iSCSI Storage Volume can be restricted to either a single machine or to multiple machines in a Subnet.

Configuring initiator group and restricting access1. In the ElastiCenter, select Accounts.


3. Select Manage iSCSI Initiator Groups in the Actions pane.

4. Select Add Initiator Group in the Tasks menu.

5. Provide the required details.

Field Description

Name Unique name of the initiator group.

175


Allowed Intiator List of IQNs, separated by comma for restricting the access to iSCSI Storage Volume frommultiple machinesin a Subnet.

Authorized Net-works

List of allowed networks, followed by subnet mask in the format 10.0.0.0/8, separated by comma.

6. Click OK.

7. In the Storage Volumes page, select the Storage Volume to which you want to assign the initiator.

8. Select Manage iSCSI Configuration (Actions pane > Tasks).

9. Click Edit, enter the required details, and then save the changes.

10. In the Confirmation dialog box, click Modify.

iSCSI authentication workflow

An iSCSI Authentication Group is a grouping of multiple iSCSI users. You can select the users who can access the iSCSI target.After you create users, you can group them as iSCSI users for authentication.

1. In the ElastiCenter, select Accounts.


176


3. Select Manage iSCSI Authentication Groups in the Actions pane.

4. Click Add Authentication Group in the following page:

177


5. Enter the required details and then click OK.

Field Description



CHAP Username Username for CHAP authentication. For uniqueness, specify the initiator IQN name as username. CHAP isan authentication method which uses a user/secret or IQNs for outgoing authentication. The StorageVolumes on CloudByte ElastiStor authenticate the initiator.

CHAP Secret Password for CHAP authentication in alpha-numeric format.The password must have 12 to 16 characters.

Mutual CHAP User-name

Username for mutual CHAP authentication. For uniqueness, specify the initiator IQN name as username.The value must be same as that of the CHAP user. Mutual CHAP is a super set of CHAP which uses peeruser/peer secret for outgoing/incoming authentication between Storage Volumes and initiator.

Mutual CHAPSecret

Password for mutual CHAP authentication (should be different from CHAP secret).

Description The password must have 12 to 16 characters.

6. In the Storage Volumes page, select the Storage Volume for which you want to assign the authentication group.

178


7. Select Manage iSCSI Configuration (Actions pane > Tasks).

8. Click Edit, enter the required details, and then click Save. In the following screen, you have selected CHAP as the authen-ticationmethod.

9. In the Confirmation dialog box, click Modify.

179


Restricting access to iSCSI VSM by managing Authentication Groups

The procedures are same as in the section Restricting access to iSCSI Storage Volume by managing Authentication Groupsexceptthat, instead of Storage Volume, you have to select VSM from step 7 onward.

Installing MPIO onWindows Server

Multipath I/O (MPIO) is aMicrosoft-provided framework that allows storage providers to developMultipath solutions that containhardware-specific information required to optimize connectivity with storage arrays.

MPIO provides an alternate data path between storage devices andWindows operating system. MPIO is protocol-independent andcan be used with Fibre Channel (FC), Internet SCSI (iSCSI) and Serial Attached SCSI (SAS) interfaces.

InstallingMPIO functionality involves the following procedures:

Note: The following documentation is forWindows Server 2008. The procedures are similar for other versions of WindowsServer. However, theremight beminor variations owing to the differences in UI strings and navigation.

1. OnWindows Server, select Multipath I/O and then click Install (Server Manager Console > Add Features Wizard).

Note: This step can be skipped if Multipath I/O is already installed.

180


2. Restart theWindows Server after you complete the installation.

3. After reboot, configureMultipath as follows:

1. OpenMPIO application (Start > Administrative Tools > MPIO).

2. Click Discover Multi-paths tab.

3. Enable Add support for iSCSI devices.

181


4. Click Add.

4. In the Reboot required dialog box click Yes.

After reboot, a new device called "MSFT2005IBusType 0x9" (or similar ID) appears in theMPIO Properties.

182


Configuring Multipath on ElastiCenter

Prerequisites:l In the Network Interface Cards page (Nodes> Network Interface Cards), configure static IP addresses for each inter-

face to configuremultipath as shown in the following screen:

183


Procedures1. In the VSMs page, click Provision VSM and then complete the procedures to create a VSM.

2. Select the provisioned VSM and then in the VSM page select Advanced Settings.

3. In the Advanced Settings page, click Edit.

4. In the following screen, provide the required details and then click Save.

Field Description

Secondary IP Address Alternate IP Address to configure multipath for a single LUN.

184


Secondary Subnet Subnet address for multipath configuration.

Secondary Interface Alternate Interface to configure multipath.

5. Provision a Storage Volume in the VSM onwhichMultipath I/O is configured.

6. (In the following page) Enter the required details and complete the procedures.

Configuring MPIO onWindows Server for a LUN in ElastiStor

Adding the first path to the iSCSI target1. Add two target portal IP addresses of the Disk Station (iSCSI Initiator Properties > Discovery).

2. Select a target that uses MPIO and then click LogOn (iSCSI Initiator Properties page > Target).

1. Select the option Automatically restore this connection when the computer starts.

2. Select the option Enablemulti-path.

3. Click Advanced.

185


3. In the Advanced Settings page, check the following options:

l Microsoft iSCSI Initiator.

l First Subnet of the Initiator Computer

l First Subnet of the DiskStation (both the IP addresses of the initiator and the target are within the same Subnet)

4. Click OK and ensure that the target has been connected.

186


5. Close the iSCSI Initiator Application.

Adding the second path to the iSCSI target1. In the iSCSI initiator properties page:

1. Select the Target to useMPIO (ensure that it is connected).

2. Click Log on.

187


2. In the Advanced Settings page, check the following and then click OK.

l Microsoft iSCSI inititator

l Second Subnet of the Initiator Computer

l Second Subnet of the DiskStation (both the IP addresses of the Initiator and the Target are within the same Subnet)

188


3. Select Details in the iSCSI Initiator Properties page.

189


4. In the Target Properties page, click Devices tab.

190


5. Select a device name and then click Advanced

191


6. Select a load balance policy and then click OK. The following screenshot shows Round Robin with Subset as the Load bal-ance policy.

192


Data access using CIFS

The Common Internet File System (CIFS), is a network protocol whosemost common use is sharing files on a Local Area Network(LAN).

The following workflow explains how to configure CIFS options:

193


Configure VSM CIFS options at the VSM level. File system CIFS options can be configured at the Storage Volume level (if you haveenabled the protocol option).

VSM CIFS options

By default, CloudByte ElastiStor creates a default VSM CIFS options group with industry standard options.

After you create the VSM, Manage CIFS Configuration option can be used tomanage CIFS options at VSM level. You can eitherselect the default one or select options from the drop downmenu.

Field Description

AuthenticationModel

By default the Authentication Model is user.

NetBios Name A unique name for the VSM CIFS options template.

Server Description A meaningful description of the VSM CIFS options template.

Workgroup The workgroup that the server will appear to be in.

DOS Charset The character set that is used while communicating with the Windows clients. The default is CP437

UNIX Charset The character set that is internally used. The default is UTF-8.

Log Level Sets the amount of log/debug messages that are sent to the log file. Default is Minimum. Keep itthis way for best performance unless more information is needed for troubleshooting.

Time Server Determines whether or not ElastiStor advertises itself as a time server to Windows clients.

File system CIFS options

By default, CloudByte ElastiStor creates a default file system CIFS options group with industry standard options. After you createthe Storage Volume, Manage CIFS Configuration option can be used tomanage CIFS options at Storage Volume level.

194


Option Description

Status If Enabled, CIFS Share can be accessed. If NFS is enabled on a Storage Volume, you can access CIFS share by settingStatus to enable.

Name A unique name for the file system CIFS options template.

Description A meaningful description of the CIFS options template.

Read Only If checked, prohibits write access to the share.

BrowseableIf selected, clients when they browse the shared directory using Windows Explorer can see the list of shares. If unse-lected, client has to explicitly specify the full server path in Windows Explorer.

Inherit Per-missions

If selected, permissions on new files and directories are inherited from parent directory.This option is (particularly) help-ful in scenarios where there are large systems with many users involved. Inheriting permissions let a single home shareto be used flexibly by each user. Leave this unselected if Type of ACL is set to Windows in the Volume's permissions.

Recycle Bin If selected, instead of purging the files, they are moved to recycle bin of CIFS share on deletion.

Hide DotFiles

Dot files (unlike on UNIX) appear on Windows. Select the option to hide the Dot files.

Hosts Allow Host-based protection feature. Comma-separated or space separated list of IP addresses that are allowed to access theshares. If nothing is specified, the default value ALL applies.

Hosts Deny Host-based protection feature. Comma-separated or space separated list of IP addresses that are denied access to thefile system. If you specify ALL, IP Addresses that are not listed in the Host Allow field are denied permission.

CIFS authentication workflow

A CIFS Authentication Group is a grouping of multiple CIFS users. You can select the users who can access the CIFS share.

1. In the ElastiCenter, select Accounts.


3. Select Manage CIFS Authentication Groups in the Actions pane.

4. Click Add Authentication Group in the following page:

195


5. Specify the required details and then click OK.

Field Description


196



User name Specify a name for the user.

Password Specify a password for the user.

Full name Specify the full name of the user.

6. In the VSM page, select the VSM for which you want to assign the authentication group.

7. Select Manage CIFS Configuration (Actions pane > Tasks).

8. Click Edit, select a Discovery Authentication Group from the drop-down list and then click Save.

197


9. In the confirmation prompt, click Modify.

Data access using FC

Fiber Channel (FC) Protocol is the standard transport protocol for delivering serial SCSI commands over Fiber Channel inter-connects. Fiber Channel Protocol is designed to operate in a highly-efficient manner using hardware for protocol offload engines(POEs). Fiber Channel is fast, reliable, and scalable data transport system. For further details, see http://en.wiki-pedia.org/wiki/Fibre_Channel.

Supported Hardware for Fibre Channell QLogic HBA is used by CloudByte as FC targets.

l The client can use any available HBAs in themarket (such as Emulex or Brocade).

For more information, see theHardware Compatibility List.

FC in CloudByte ElastiStor

The following figure illustrates how you configure the FC setup in CloudByte ElastiStor:

Access over FC protocol involves an initiator group and a target group. Initiator represents the FC client, for example aWindows cli-ent and target is the CloudByte Node.

198

http://en.wikipedia.org/wiki/Fibre_Channel



The Storage Volumes can be accessed by both the initiator group and the target group.

FC Initiator group

An initiator group is the list of initiators who are authorized to access the FC LUN. By default, CloudByte creates a default initiatorgroup named none that does not allow access to any initiator.

The initiator group is a list of WWPNs. You can create the initiator group, add initiators and ports, and thenmask the LUNs to theadded initiator.

You create the initiator group at the account level and assign it at the Storage Volume level.

FC Target group

You group the FC target ports to form a target group. This means, all the FC target ports which you have grouped are used for thisparticular group. The Node’s WWPNs uniquely identify each target port on the system. By default, CloudByte creates a default tar-get group named ALL that allows access to all targets. If required, you can create a new target group and replace the default one,though CloudByte recommends using the default target group ALL.

You create the target group at the account level and assign it at the Storage Volume level.

Creating an initiator group1. In ElastiCenter, select Accounts.


3. In the Action Pane, select Access Protocol Options > Manage FC Initiator Group.

4. Refresh Hardware (Nodes > select the specific Node> click Refresh Hardware in the Tasks section).

Note: Step 4 applies only if you have added FC ports to the Node after adding the Node to the HA Groups.

5. In the following page, click Add Initiator Group.

199


6. Specify a unique name and select the appropriateWWPNs of the FC initiators to form the group.

7. Click Add.

Creating a target group1. In ElastiCenter, select Accounts.


3. In the Action Pane, select Access Protocol Options > Manage FC Target Group.

4. Refresh Hardware (Nodes > select the specific Node> select Refresh Hardware in the Tasks section of the task list).

Note: Step 4 applies only if you have added FC ports to the Node after adding the Node to the HA Groups.

5. In the following page, click Add Target Group.

6. Specify a unique name andWWPNs to form the group.

7. Click Add.

Assigning Storage Volume to initiator group and target group1. In the Storage Volumes page, select the Storage Volume for which you want to assign the initiator/target group.

2. In the Action Pane, select Tasks > Manage FC Configuration.

200


3. In the FC Configuration page, click Edit and then replace the default initiator group/target group.

4. Click Save.

Note:You can also perform the same task at the Account level (Accounts > select Account > Manage FC Initiator Group/Tar-get Group).

Data access using NFS

Network File System (NFS) allows a server to share directories and files with clients over a network

Note: This release does not support access control when using NFS.

The following section details the setup:

VSM NFS options1. In the VSM Page, click Manage NFS Configuration (Actions > Tasks).

2. (In the following page) Click Edit, change the Number of NFSWorker Threads, and then save the settings.

File system NFS options1. In the navigation pane, select Storage Volumes.

2. In the Storage Volume page, select the Storage Volume for which you want to specify the NFS options.

201


3. In the following page, select Manage NFS Configuration in the Tasks section of the actions pane.

4. In the Authorized NFS clients page, click Add NFS Clients.

5. Specify the following NFS details:

Option Description

Network/IPaddresses

Single Network ID with subnet mask (for example,10.10.1.100/8). You can also provide a comma-separatedlist of IP addresses for which access has to be provided (for example,10.20.10.120, 10.20.10.121) instead.

All Dir-ectories

Yes the client can mount any sub directory within the Path.

Map as RootUsers

l Yesrestricts all users to root.

l No restricts write or delete accesses to other users' files.

Read-only Yes prohibits writing to the share.

6. Click Add.

202


Restore snapshots while mapping iSCSI LUNs

On ESX Server1. In ElastiCenter, create a Storage Volume.

2. Mount the Storage Volume on ESX client:

1. (In the following screen) Click on ESX Server and then select the configuration option.

2. In the iSCSI initiator properties window (Storage Adapters > Properties), select Dynamic Discovery.

3. Select Add (Add Send Target Server window appears).

4. In the iSCSI tab, specify the tenant IP Address and then click OK.

3. Add Storage.

1. Click Storage in the left pane.

2. Select Add Storage.

3. Mark the storage type as Disk/LUN and then click Next.

203


4. (In the following screen) Select the listed LUN and then click Next.

5. Mark the File System Version as VMFS-5.

6. Specify a name for the DataStore and then click Next.

7. (In the following page) Mark Maximum available space option and then click next.

8. Click Finish.

4. Copy data to the new Volume.

204


1. Right Click on the Datastore name and then select Browse Data.

2. Click on the upload icon to copy data as seen in the following page:

5. Create a snapshot on the Storage Volume using the following procedures:

1. In the Storage Volumes page, select the Storage Volume for which you want to create a local backup.

2. In the following page, select Create Snapshot in the Actions pane.

3. Specify a name for the snapshot to be created and then click OK.

6. Delete data from the DataStore on the ESX Client.

205


7. In ElastiCenter, click Manage Snapshots (Actions pane > Local Backup) to view the list of previously captured snapshots.

8. From the list of snapshots, select the snapshot you want to restore and then click Restore icon to roll back the StorageVolume content to previous state.

9. On the ESX client, refresh the Storage Volume and then view the required data.

OnWindows XP, Windows 7, Windows 8, Windows Server 2008, andWindows Server 20121. In ElastiCenter, create an iSCSI Storage Volume.

2. Mount the Storage Volume on theWindows Client.

1. OnWindows machine, launch iSCSI Initiator (Programs > Administrative Tools > ISCSI Initiator).

2. In the Target tab, specify the target (tenant IP address).

206


3. Click Quick Connect. TheQuick Connect screen appears, displaying the target name.

4. Select the Target name and then click on Connect.

5. Click OK.

3. Create Storage Drive to use the iSCSI target.

1. Launch Server Manager and then select Disk Management in the left pane.

2. Right-click the unallocated space (New Simple Volumewizard appears).

3. Select New Simple Volume.

207


4. Ensure that the new Storage Volume appears as shown in the following page.

5. Copy data to the new Volume.

208





3. Specify a name for the snapshot and then click OK.

7. Delete the data on themounted Storage Volume from theWindows Client.


209


9. Before restoring the snapshot, make the volume offline on theWindows Client (Disk Management > Right-Click Disk).


11. On theWindows Client, bring the Storage Volume online.

Restore snapshots while mapping CIFS share

On Windows

1. In ElastiCenter, create a Storage Volume and name it, say, testvol1.When you create the first CIFS Storage Volume, the Create User page appears.

2. Create user credentials to mapCIFS share onWindows Client as follows:

210


1. In the Accounts page, select the Account associated with the Storage Volume.

2. Select Manage CIFS Authentication Groups in the Actions pane.

3. Select Add Authentication Group in the Tasks section.

4. In the following page, provide the relevant credentials and then click OK.

211


3. Mount the Storage Volume onWindows Client as follows:

1. Open yourWindows Client and then select Computer > MapNetwork Drive.

2. Write some data.

4. Ensure that you have enabled CIFS as follows:

1. OnWindows Client, launchMapNetwork Drive dialog box (Computer > MapNetwork Drive).

2. Specify the drive and the tenant IP address (with CIFS share name) in a format as shown in the following screen:

3. Provide the CIFS share access path and then click Finish.

4. Ensure that the option Connect using different credentials is checked.

5. Provide the CIFS username and password and then click OK.

6. Ensure that CIFS share is connected (highlighted), as shown in the following page:

212


7. Copy some data.





9. Go toWindows Client and then delete the data from themounted Storage Volume.


11. For the snapshot you want to restore, click Restore icon to roll back the Storage Volume content to previous state.

213


12. OnWindows Client, refresh the Storage Volume and then check the data.

Restore snapshots while mapping FC LUNs

On ESX Server1. In ElastiCenter, create an FC Storage Volume.

2. In ElastiCenter assign an ESX Server as initiator to the FC Storage Volume.

3. (In the following screen) Click on ESX Server and then select the configuration option.

214


4. Add Storage:

1. Click Storage in the left pane.

2. Mark the storage type as Disk/LUN and then click Next.

3. (In the following screen) Select the listed LUN and then click Next.

4. Mark the File System Version as VMFS-5.

5. Specify a name for the DataStore and then click Next.

6. (In the following page) Mark Maximum available space option and then click next.

7. Click Finish.

5. Copy data.

215


1. Right Click on the Datastore name and then select Browse Data.

2. Click on the upload icon to copy data as seen in the following page:





7. Delete data from the DataStore on the ESX Client.

216




10. On the ESX client, refresh the Storage Volume and then view the required data.

OnWindows Server 20081. In ElastiCenter, create an FC Storage Volume.

2. In ElastiCenter, assign aWindows client as an initiator to the FC Storage Volume.

3. Create Storage Drive to use the FC LUN.


2. Right-click the unallocated space (New Simple Volumewizard appears).

217


3. Select New Simple Volume.

4. Ensure that the new Storage Volume appears as shown in the following screen:

5. Copy some data to the Volume.

218





3. Specify a name for the snapshot and then click OK.

7. Delete the data on themounted Storage Volume from theWindows Client.


219


9. Before restoring the snapshot, make the volume offline on theWindows Client (Disk Management > Right-Click Disk).


11. On theWindows Client, bring the Storage Volume online.

Restore snapshots while mapping NFS Share

On ESX Server1. In ElastiCenter, create a Storage Volume.

2. Mount the Storage Volume on ESX Client.

1. In the following page, Click on ESX Server.

2. Click Storage in the left pane under the Hardware section.

220


3. Select Add Storage option.

4. Mark the storage type as Network File System and then click next.

3. Add Storage.

1. In the Server tab, specify NFS Server IP Address.

2. In the Folder tab, specify the access path of NFS share.

3. Specify a name for DataStore and then click Next.

4. Copy some data.

221


1. Right Click on the Datastore name (Browse Datastore wizard appears).

2. Select Browse Datastore.

3. Click on upload icon to copy data as seen in the following page:





6. Delete the data from the DataStore on the ESX Client.

222




9. On ESX Client, refresh the Storage Volume and then view the required data.

On Linux Client1. In ElastiCenter, create an NFS Storage Volume.

2. Mount the volume on Linux client and then write some data.Use the following command to mount the volume: mount Tenant IP Address : Access Path Directory Path

l Directory Path: Local Directory Path where the NFS volume is to bemounted.

Example: mount 20.10.17.117:/Acctestvol3 /mnt/vol1

To copy data to the volume, use the following command:cp /Source Filename Destination Filename

Example: cp /root/Desktop/test2 /mnt/vol1cp /root/Desktop/test1 /mnt/vol1

223






4. Delete the data from themounted Storage volume using the following command:rm -r Filename FilenameExample: rm –r test test2


6. From the list of snapshots, click on Restore icon to roll back the Storage Volume content to previous state.

7. On the Linux Client, check data. Use the ls command to view the list of files.

224

CHAPTER VIII Managing CloudByte ElastiStor


The following sections explain themanagement features in CloudByte ElastiStor:

n "Delegated administration"

n "Monitoring""Monitoring"

n "Drill-down feature"

n "CloudByte ElastiStor reporting"

Delegated administration

Administration in CloudByte can be categorized based on services. You can have administration at the Site level or at the Accountlevel. Your Site can havemultiple Site administrators.

Note: Currently, CloudByte ElastiStor supports only role-based administration. Future releases will have delegated admin-istration fully implemented.

As the root administrator, you can distribute the tasks to administrators. Delegated administration is an act of providing and restrict-ing privileges by establishing access control based on the responsibility of each individual administrator.

n Delegated administration lets you determine the actions that each administrator can take using administrative tools on dif-ferent resources.

n Delegated administration roles aremapped to administrative functions using security policies. You can create andmaintainseparate administration capabilities, depending on your needs.

n You establish delegated administration by creating role hierarchy.

n Role hierarchy enables you to keep a tight control on how delegation happens and who can delegate what to whom. Oneadministrator can create sub-roles for other administrators with limited administrative privileges.

Role hierarchy

The following is a pictorial representation of how role-hierarchy functions in CloudByte ElastiStor administration:

225


Administrator Can Cannot

Super admin-istrator

Perform all possible actions across Sites, Accounts, and VSMs Not applicable

Site administrator l Manage permissible actions within the permitted Sites

l Manage more than one Site

l Have only read-only access

l Create/Delete/Modify Accounts (if permitted by superadministrator)

Delete/Modify/Create another Site admin-istrator

Account superdministrator

l Manage only that Account

l Do everything permissible for an Account administrator

l Manage anything outside the Account

l Manage anything with respect to theinfrastructure

Account admin-istrator

Manage what is permitted within the Account Manage other account admins within theAccount.

226


View admin-istrator

l View all the permitted Sites and underlying infra-structure

l Have only read-only access

l View all the permitted Accounts

l View Cluster administrators and Account administrators

l Generate reports

l Delete/Modify/Create another Viewadministrator

l View Site administrators

Note: You can only have one super administrator. Only the super administrator can change his/her password. Admin-istrators can change their password and passwords of administrators under them.

Adding delegated administrators

Create Site administrators or view administrators

Note: You can create Site administrators or view administrators only if you are a super administrator.

1. Log in as a super administrator.

2. In the navigation pane, select Delegated Administrator. The Delegated Administrators page appears.

3. Click Add Delegated Admin.

4. In the Add Delegated Admin Details page, specify the following details:

227


l Email: Specify a unique email address for the Site administrator or View administrator

l Password: Administrator password

l Delegated Admin Type:Select the type of Delegated Administrator from the drop-downmenu

5. Click Next and then drag and drop one or more Sites for which you want to delegate administration as seen in the followingscreen:

6. Click Add.

228


Create Account administrators

You can create Account administrators for either existing or new accounts.

Create Account administrators for a new account

Note: The type of administrator that you can assign tasks to depends on the privileges you have.

1. In ElastiCenter, select Accounts in the navigation pane.

2. In the Accounts page, click Add Account.

3. In the Account Details page, when you specify the account details, choose the type of administrator that you want to create(Account Super Admin or Account Admin).

4. Specify the credentials for the administrator and then click Add Account.

Create Account administrators for an existing account

Note: The type of administrator that you can assign task to depends on the privileges you have.

1. In ElastiCenter, select Accounts in the navigation pane.

2. In the Accounts page, select the account for which you want to add new administrators.

3. In the following page, click Add Administrator in the Delegated Administration section of the tasks pane.

229


4. In the Add Account dialog box, specify the details and then click OK.

Monitoring

Monitoring provides an intuitive visualization of storage statistics.

Monitoring foregrounds relevant backend activities in an easy user interface for administrators. It provides a generic outlook of thecomponents such as storage, input/output, throughput, latest data, status, alerts, and system configuration.

Various dashboards dynamically represent disk utilization, storage utilization, throughput, input/output, and status of network inter-face cards attached to the system.

Alerts and Events feature keeps a regular check on what is happening with the product. All events - even adding an additional IPaddress or modifying an existing one - get registered andmonitored.

The feature also provides a notification system where the administrator is informed through email/text messages when an alert isgenerated in the system.

Continuous monitoring of your network

n Helps you discover problems long before they become critical

n Enables you to take proactivemeasures

n Ensures that performance is always at expected levels

n Ensures that resources are effectively utilized

Monitoring in CloudByte ElastiStor

Monitoring in CloudByte ElastiStor can be classified as follows:

n Node level alerts: The alert information specific to the Node.

n Dashboard alerts: Graphical representation of monitoring statistics at granular levels.

230


n Events: A snapshot of all ElastiCenter-specific events.

Node level alerts

Node level alerts provide a list of alerts specific to the Nodes.

1. In ElastiCenter, click Alerts. All alerts associated with the Node are displayed.

2. Click the Acknowledge icon to acknowledge the alert and it disappears from the list. To acknowledge all the alerts, click BulkAcknowledge in the Actions pane.

Emailing alerts

The following settings in Global Settings let you email alerts:

l By default , alerts are not emailed. To receive alerts over email, set alert.email.enable to true . The alerts aremailedto the email that you specified when you updated your profile. For details, see "Reset password and configure auto-alertoptions". To addmultiple users, add the email IDs separated by comma in alert.email.addresses.

l You can set a filter to the alerts mailed based on severity using the option alert.email.severity. By default, all alertsaremailed (as the value set is 4).

l To specify the sender email ID, use the option alert.email.sender. If you do not specify an ID, the ElastiCenter domainname is displayed along with IP address of ElastiCenter.

Events

To get a list of ElastiCenter events,

1. In ElastiCenter, click Events.

2. Select Events in the Select View. All events are listed.

To get a list of specific events, go to the component page and then click Events. For example, for events related to only StorageVolumes:

1. In the Storage Volumes page, select a Storage Volume for which you want to generate events.

2. In the following page, click Events in the actions pane.

231


Dashboard Alerts

Dashboard provides a visual representation of monitoring statistics such as performance, usage, capacity, and availability.

To see the details, select the component and thenmake a specific selection.

The following figure is a dashboard representation of the alerts

Advanced debugging options

The Troubleshooting option provided in the Node page, Pool page, and VSM page provides a set of information that help you withadvanced debugging. The options provide information such as the following:

l For Pools: Pool I/O Status and Pool Properties

l For VSMs: NIC and Netstat information

l For Nodes: I/O Status of various disks, GMultipath status, and Camcontrol device list for all available disks

When you select an option, details are provided in a tabular format. To access the Troubleshooting option,

1. In the Node/Pool/VSM page, click a specific Node/Pool/VSM.

2. In the Actions pane, click Troubleshooting under Tasks.

232


3. Select the action and then click Update.

233


Drill-down feature

A monitoring feature in Pools, Storage Volumes, and VSMs that lets youmonitor

n IOPS

n Throughput

n Latency

n Average block size (applicable only to Storage Volume)

n READ vs WRITE (applicable only to Storage Volume)

Select a time interval option such as start time and end time, last 24 hours, or last week to plot a graph.

Using the drill-down feature

To use the drill down feature, select the component and then select a specific feature.

Drill-down procedure for Site

1. From the list, select a Pool for which you want to monitor IOPS.

2. Click Compare Volumes.

3. In the following page, set the desired parameters.

234


4. Click Update.

Comparing IOPS

1. From the list, select the Pool for which you want to monitor Storage Volume IOPS.

2. Click Compare Volumes.

3. In the following page, set the desired parameters.

4. Click Update.

Note: IOPS comparison is applicable only for Storage Volumes in the same Pool.

CloudByte ElastiStor reporting

CloudByte ElastiStor provides detailed reports that you can save or print. The reports can be generated in the formats PDF, HTML,and CSV. You can schedule reports and email the generated reports.

You can generate reports at granular levels, for instance at Account level or Pool level. Reports help you with various statistics suchas QoS data.

CloudByte provides the following reports, which you can generate, print, and circulate:

l Account IOPS Provisioning

l Pool IOPS Provisioning

l Account Storage Provisioning

l Pool Storage Provisioning

l Account QoS

l Pool QoS

235


l Account Storage

l Pool Storage

l VolumeUsage

l HA Group Info

l Global Accounts IOPS Storage Provisioning

Note: In future releases, you can define your own reports and add new report types.

Accessing the reports1. In ElastiCenter, select Reports.

2. Click the respective icon to generate or schedule a report.

3. In the dialog box that appears, specify the relevant details and then click OK.

Configuring email for reports

To send the generated reports over mail, Click Generate new Report icon in the Actions pane and then specify the address to whichthe report has to bemailed.

236


To change the default settings for configuring email for reports, select Global Settings and then specify the details for the highlightedoptions:

237


Global Settings

Global settings let the administrator configure various options used to control CloudByte ElastiStor. For example, email con-figuration. To configure global settings,


2. Select a target category. A list of settings is provided by default.

3. Edit the relevant settings.

Note: You cannot add new settings.

This section lists the global settings options available on the ElastiCenter.

Option Description Default Value

alert.email.addresses Comma separated list of email addresses usedfor sending alerts.

Email ID you specified in the UpdateProfile section in ElastiCenter.

238



alert.email.enable If true, emails are sent to the specified emailaddresses.

TRUE

alert.email.sender Sender of the alert email.

alert.email.severity Set the email severity level to one of the fol-lowing:

l 1 - ERROR

l 2 - CRITICAL

l 3 - WARNING

l 4 - INFO

4

alert.smtp.host SMTP hostname to send alert emails. localhost

alert.smtp.password Password for SMTP authentication (applies onlyif alert.smtp.useAuth is true).

alert.smtp.port The port of the SMTP server.587

alert.smtp.useAuth If true, SMTP authentication is enabled to sendemails.

False

alert.smtp.username Username for SMTP authentication (applies onlyif alert.smtp.useAuth is true).

autoSupport.config.details Send configuration details to CloudByte as partof an auto support email.

Yes

autosupport.httpnotify.apikey API key of CloudByte auto support server.

autosupport.httpnotify.enable Send HTTP post request to CloudByte auto sup-port server. False

autosupport.httpnotify.ipaddress The IP address of CloudByte auto support server.

autoSup-port.infastructure.statistics

Send infrastructure statistics to CloudByte aspart of an auto support email.

Yes

autoSupport.product.alerts Send product related alerts to CloudByte as partof an auto support email.

Yes

autoSupport.reg.details Send registration details to CloudByte as part ofan auto support email. Yes

autoSupport.tenant.statistics Send tenant statistics to CloudByte as part of anauto support email.

Yes

bad.disk.threshold The number of defects grown on disk that willdeclare it as a bad disk

512

cache.multiplication.factor The number of Cache READs which equal a diskREAD. The default value is 3 which means 1 diskREAD equals 3 cache READs. Only a valuebetween 2 and 6 is supported.

3

devman.backup.files.retain The number of backup files to be retained. 5

devman.backup.path The path of ElastiCenter backup directory. /var/devman/backups

239



devman.backup.scheduling.interval The interval time for ElastiCenter backup. 1

devman.build.number The build version of ElastiStor. Available in /cbdir/buildversion

monitor.purge.retain.days The number of days that the data has to beretained for.

60

monitor.qos.stats.interval The time interval (in seconds) taken for QoS stat-istics to be retrieved from Nodes.

300

monitor.refresh.interval The time interval (in seconds) required torefresh the Storage Volumes page.

300

monitor.storage.stats.interval The time interval (in seconds) taken for storagestatistics to be retrieved from Nodes.

300

monitor.system.purge.retain.days The number of days that the data has to beretained for.

7

no.of.nodes.per.hagroup Limiting number of Nodes in an HA Group.

project.invite.required Specify if invitation confirmation is requiredwhen add account to project.

False

provisioning.sb.algo Algorithm used for auto-provisioning StorageVolume.

Best Fit

provisioning.sb.auto Enable/Disable auto-provisioning of StorageVolumes.

No

provisioning.tsm.algo Algorithm used for auto-provisioning VSM. Best Fit

provisioning.tsm.auto Enable/Disable auto-provisioning of VSMs. No

qos.penalty.allowed.delta Percentage (as a value between 1 and 100) up towhich penalty will be not be considered

10

qos.penality.blocksize Standard blocksize (in KB) per IOPS, used in pro-visioning and penality enforcement.

4

qos.penalty.blocksize.factor Considers block size for penalty calculations. Spe-cify yes to Enable and No to Disable.

No

qos.penalty.cachehit.factor Considers cache hit for penalty caclulations. Spe-cify yes to Enable and No to Disable.

No

qos.pen-alty.enforcement.timeinterval

The time interval (in seconds) for which penaltyhas to be applied on QoS values.

300

qos.penalty.readwrite.factor Considers READs vs WRITEs for penalty caclu-lations. Specify yes to Enable and No to Disable.

No

read.multiplication.factor Specify howmany READs equal a WRITE. Thedefault value is 4 which means 1 WRITE equals 4READs. Only a value between 3 and 6 is sup-ported.

4

report.email.sender Sender of report email.

report.smtp.host SMTP hostname used for sending out reportemails.

localhost

report.smtp.password Password for SMTP authentication (applies onlyif report.smtp.useAuth is true).

240



report.smtp.port The port of the SMTP server. 587

report.smtp.useAuth If true, SMTP authentication is enabled to sendemails.

False

report.smtp.username Username for SMTP authentication (applies onlyif report.smtp.useAuth is true).

ui.protocol.enablefc Enable/Disable all FC protocol related optionsfrom ElastiCenter.

True

ui.protocol.enablehaadmincreation Enable/Disable haAdmin creation options fromElastiCenter.

False

ui.protocol.enablezilmirroring Enable/Disable ZIL mirroring creation optionsfrom ElastiCenter.

False

provisioning.uncontrolled.iops Enable/Disable provisioning with uncontrolled(zero) IOPS. Specify TRUE to enable.

False

alert.qos.throttling.percentage Sends an alert if the throttled IOPS percentageexceeds the specified value and if aler-t.qos.throttle.count is encountered for thespecified number of times.

20

alert.qos.throttle.count Sends an alert if the number of consistent throt-tling exceeds the specified value and aler-t.qos.throttling.percentage exceeds thespecified value.

10

devman.enable.ipspaces Enable/Disable the Assigning real IPspaces feature.

False

241

CHAPTER IX Licensing


When you install the Community Edition of CloudByte ElastiStor, a free perpetual license file is already uploaded to ElastiCenter.You can use the license for up to 25TB usage of storage on CloudByte ElastiStor.

The license file has relevant details that implement license to use various features that you have purchased. CloudByte provideslicense file in the format L4J.

The following FAQs help you with licensing.

How do I access license details?1. Click Admin (at the top-right corner of the page).

2. Select License from the drop-down list.

3. The License page appears.

I want to evaluate CloudByte ElastiStor. Howwill the trial license be enforced?

When you install a the Community Edition of CloudByte ElastiStor, the free perpetual license file is, by default, uploaded. You neednot do any licensing tasks.

I received the license file from CloudByte. Howwill I proceed?1. Place the license file (with the extension .l4j) on themachine from which you access CloudByte ElastiCenter.

2. Log in to CloudByte ElastiCenter.

3. In the Actions pane, click Upload New License and then Upload License.

242


The Add License wizard appears.

4. Choose the license file that you have received and then click Upload.

The Check screen confirms the validity of the license and the Finish screen provides the license details. When you get backto the License page, all license details are listed.

Can I check the past licensing activities?

In the License page, click View History. The details of the license are listed.

243


Can I get support from CloudByte for the trial version of the product?

Yes. Contact CloudByte technical support at [email protected]

244

http:// [email protected]/

CHAPTER X CloudByte ElastiStor REST APIs


CloudByte ElastiStor REST APIs provide a bundle of functions that encompasses all the features in the CloudByte ElastiStorAdministrator Console (CloudByte ElastiCenter). You do not need ElastiCenter to configure andmanage CloudByte ElastiStor.Rather you can use your own portal.

The CloudByte ElastiStor REST APIs provide powerful, convenient, and simple REST-based APIs which can interact withCloudByte ElastiStor.

Click one of the following links to get a complete list of CloudByte ElastiStor REST APIs and documentation relevant to your admin-istrative privileges:

l Super administrator APIs

l Site administrator APIs

l Account super administrator APIs

CloudByte ElastiStor API references

The following sections explain the implementation.

How to use CloudByte ElastiStor REST APIs1. Install CloudByte ElastiCenter. For details, see "Installing CloudByte ElastiStor".

2. Authenticate as Administrator. For details, see "Launching CloudByte ElastiCenter"

3. Obtain the CloudByte REST API authorization token. This token is passed in each request using the API Key request para-meter. For details, see "Obtaining the REST API key".

4. Form the request URL. For details, see "Forming the API request".

Obtaining the REST API key1. Log in to ElastiCenter as Super Admin.

2. In the navigation pane, select Delegated Administrator. The list of administrators appears.

3. From the list of Accounts, click the specific account for which you want to generate the API key.

4. In the following page, click Generate Keys.

245

http://www.cloudbyte.com/elastistorapi/TOC_Root_Admin.html

http://www.cloudbyte.com/elastistorapi/TOC_Domain_Admin.html

http://www.cloudbyte.com/elastistorapi/TOC_User.html


5. Click Yes to confirm. The API key is generated and displayed as shown in the following screen:

Forming the API request

The following is a sample GET request:

https://<ElastiCen-terIPaddress>/client/apicommand=listSite&apiKey=Xxj48vo357IsQ5H5KOPQksMzFBGYgTJNqgItBAulgu5YeOkwT4L8DAWUb_p43Nlv23e4eBeuN3Ts8_0-m3AyQA&response=json

Interpreting the URL

URL Component Description

246


https://<ElastiCenter IP address>/client/api? Base URL which includes the IP address to accessCloudByte ElastiCenter; the web service API entry point.

command=listSite The web service command you execute. For the com-plete list of commands, see ElastiStor APIs.

apiKey=UjGmN7gGwzMhFsSese3TNNOWiXHGn53eon_7sHx3aL-9_t3sHAOV_rNqVzwTnrJyCaTAg9FucSBvMp49nbnfEQ

The API key that you generated.

response=json Response format. This can be XML or JSON. To get anXML response, replace json with xml in the URL.

Apart from these, the URL also has the additional parameters (required or optional) for the command.

Sample response in JSON{ “listSiteResponse” : { “count”:2 ,”site” : [ {

“id”: “66291daa-4c40-4462-8747-926ebc488963”,

“name”: “Bangalore”,

“location”: “India”,

“managedstate”: “Available”,

“revisionnumber”: 0,

“currentUsedSpace”: 29,

“currentAvailableSpace”: 9943,

“currentTotalSpace”: 9972,

“currentThroughput”: 605,

“currentIOPS”: 151,

“currentLatency”: 163,

“currentThrottle”: 20,“controllers”: [

{

“nodePools”: “4”,

“id”: “4c6e2246-52c1-3227-b51a-9928cf3b675a”,

“clusterid”: “88bf812b-5142-4780-8ad3-13178bbf30a1”,

“name”: “host1”,

“hostname”: “Both”,

“nodeTsms”: “4”,

“ipaddress”: “10.36.36.50”

}

],

“haclusters”: [

{

“id”: “88bf812b-5142-4780-8ad3-13178bbf30a1”,

“haClusterPools”: “4”,

“haClusterNodes”: “1”,

“name”: “Cluster1”,

247

http://elastistor.info/


“haClusterTsms”: “4”

}

],

“hapools”: [

{

“controllerId”: “4c6e2246-52c1-3227-b51a-9928cf3b675a”,

“id”: “38e66579-bb91-3de3-815a-48c35e2c5943”,

“availiops”: “50”,

“totalthroughput”: “102400”,

“name”: “Pool1”,

“poolTsms”: “1”,

“availthroughput”: “51200”,

“totaliops”: “100”

},

{


“id”: “226e2e02-a3e5-3609-b70d-aea8ef655f74”,



“name”: “Pool2”,




},

{


“id”: “810f428d-9c26-3d91-8675-62cc3ca56da0”,



“name”: “pool3”,




},

{


“id”: “11d6fd85-5782-39fe-b85d-079d73742757”,



“name”: “P4”,




}

248


],

“tsms”: [

{


“id”: “1cb805fe-dd22-357e-b198-121fa08d0971”,



“name”: “T1”,

“ipaddress”: “10.37.37.51”,



},

{


“id”: “90a4e6f8-cb3c-317f-8cdf-3986e8759747”,



“name”: “T2”,

“ipaddress”: “10.37.37.52”,



},

{


“id”: “cab14af0-84b6-3761-900b-9675b32b6ba5”,



“name”: “T3”,

“ipaddress”: “10.37.37.53”,



},

{


“id”: “17b17719-137d-3f95-9f4f-468a17a45957”,



“name”: “T4”,

“ipaddress”: “10.37.37.54”,



}

]

}, {

249


“id”: “74692191-75a9-4294-9b1e-402127c9e327”,

“name”: “test”,

“managedstate”: “Available”,

“revisionnumber”: 0,

“currentUsedSpace”: 0,

“currentAvailableSpace”: 0,

“currentTotalSpace”: 0,

“currentThroughput”: 0,

“currentIOPS”: 0,

“currentLatency”: 0,

“currentThrottle”: 0

} ] } }

Sample response in XML<listSiteResponse cloud-stack-version=“1.1.0.4.2013-02-06T10:19:20Z”>

<count>1</count>

<site>

<id>ad63d28a-9de9-443f-8260-4309337f40c3</id>

<name>site</name>

<managedstate>Available</managedstate>

<revisionnumber>0</revisionnumber>

<currentUsedSpace>0</currentUsedSpace>

<currentAvailableSpace>1996</currentAvailableSpace>

<currentTotalSpace>1996</currentTotalSpace>

<currentThroughput>0</currentThroughput>

<currentIOPS>0</currentIOPS>

<currentLatency>0</currentLatency>

<currentThrottle>0</currentThrottle>

</site>

</listSiteResponse>

250

CHAPTER XI Connecting to ElastiStor


The following sections explain how youmap volumes to various clients:

l "Mapping iSCSI LUN"

l "Mapping CIFS share to the clients"

l "Mapping FC LUN to the clients"

l "Mapping NFS share to clients"

Mapping FC LUN to the clients

On VMware vSphere client1. Ensure that you have enabled FC for the particular Storage Volume. For details, see "Data access using FC ".

2. In the vSphere Client, go to Configuration > Storage.

3. Click Devices and then select Rescan All.

4. In the Rescan dialog box, click OK. The refreshed list of devices appears.

251


5. Click Datastore.

6. In the datastore list, select the datastore and then click Add Storage.

The Add Storage wizard appears.

252


7. Follow the instructions on theWizard and then complete the procedures to add storage.

Mapping FC LUN on Windows Server 2008

Discovering the FC target

Note: Themapping procedures are similar for other versions of Windows Server. However, theremight beminor variationsowing to the differences in UI strings and navigation.

Note:Microsoft Multipath I/O (MPIO) takes care of presenting a pseudo device.

1. Ensure that you have enabled FC for the particular Storage Volume. For details, see "Data access using FC ".

2. OnWindows Server 2008, select Server Manager > Disk Management > Rescan Disks.

253


After rescan, the Volumes that you have created for FC are discovered.

254


3. Right-click the unallocated space and then click Online.

4. Once online, right-click and then select Initialize Disk.

255


The Initialize Disk dialog box appears.

5. Click OK.

6. Right-click the unallocated space and then select New Simple Volume as shown in the following screen:

256


The New Simple VolumeWizard appears.

257


7. Follow the instructions and complete the wizard to create the partition.

8. Ensure that the created disk is added to the Disk management console (indicated by the blue ribbon).

258


9. Ensure that the drive created out of FC LUN is shown (as in) the following screen:

259


Mapping FC LUN as a raw disk1. On theWindows Virtual Machine, right-click and then select Edit Settings.

2. In the Virtual Machine Properties page, click Add.

260


3. Select Hard Disk as the device type and then click Next.

4. Select Raw devicemappings and then click Next.

5. From the list of LUNs, select a LUN which has to bemapped as a raw device.

6. Follow the wizard and complete the procedures.

7. When the Raw mapped LUN message appears, click OK.

On Kernel-based Virtual Machine1. Do either of the following:

l As root, make an SSH connection to the KVM client.

l On the KVM client, log in as root.

2. At command prompt, do the following tomap FC LUN to KVM.

3. Run the following command to verify the number of FC Hosts: ls /sys/class/fc_host/The following is a sample output:

host0 host1 host2 ..........hostX

Note: The number that follows host(1, 2, or X) is the HBA number of the FC Host.

4. Run the following command to discover the FC LUNs:

echo "1" >/sys/class/fc_host/hostHBA number/issue_lipFor example,

echo "1" >/sys/class/fc_host/host7/issue_lip

5. Repeat the step 4 for all the FC Hosts.

6. Run the following command to fetch theWWPN of the FC Host:

cat /sys/class/fc_host/host<HBA number>/port_nameFor example,

261


cat /sys/class/fc_host/host7/port_name

7. Run the following command to add each FC LUN as a device to the KVM client:

echo "scsi add-single-device h c t l" > /proc/scsi/scsi

Parameter Description

h HBA number of the Host.

c Channel ID on the HBA.

t SCSI target ID.

l FC LUN ID.

8. Run the following command to to list the disks: fdisk -l

Mapping iSCSI LUN

Note: Do not present the same LUN tomultiple clients/initiators until an OS Clustering (for example, Red Hat Cluster Suite(RHCS) or Microsoft Cluster Server (MSCS)) uses the shared LUN.

On Linux

The following procedures might work onmost of the Linux setups. However the certified versions are

l OpenSUSE-11.4-x86 -64 bit and 12.1 version

l RedHat Enterprise Linux Server release 6.0

l CentOS-6.4-x86

Without CHAP authentication

1. At command prompt, run the following command, to discover the iSCSI target: iscsiadm -m discovery -t st -p<VSM_IP_address:port>.

The iSCSI target (iqn) is displayed as shown in the following example: 10.10.8.201:3260,1 iqn.2012-09.Cer-tStorage:isci-on-iscsiSet

2. Connect to the target in either of the following ways:

l To connect to all targets, run the following command: iscsiadm –m node –l

l To connect to an individual target, run the command as shown in the sample command: iscsiadm –m node –T<target_name_displayed_when_you_run_discover_command> –p <VSM_IP_address:port> –l

A success message is displayed.

2. To createmultiple partitions from the LUN, run the following command: fdisk -l.

All available disks and partitions are listed.

3. Use the following command to create the partition: fdisk <disk(Discovered LUN)>

4. (Optional) Enter the command m to list the commandline help.

5. Create a partition using the command n (primary or extended).

6. Specify the cylinder size and value.

262


7. Specify the size of the partitions and enter w forWRITE changes.

8. Create a file system on the created partitions using the following command: mkfs.<ext2,ext3,ext4,ntfs,vfat><partition>

9. Create an empty directory using the following command: mkdir <directoryname>

10. Mount the partitions to the empty directory using the following command: mount <partition name> <dir-ectoryname>

11. Perform READ andWRITE operations onmount point.

Sample CLI snapshot

<codedoc code>[root@qaprod ~]# iscsiadm -m discovery -t sendtargets -p 10.10.8.201

10.10.8.201:3260,1 iqn.2012-09.CertStorage:isci-on-iscsiSet

[root@qaprod ~]# iscsiadm -m node -T

iqn.2012-09.CertStorage:isci-

on-iscsiSet --portal 10.10.8.201:3260 --login

Logging in to [iface: default, target: iqn.2012-09.CertStorage:isci-

on-iscsiSet,

portal: 10.10.8.201,3260]

Login to [iface: default, target: iqn.2012-09.CertStorage:isci-on-iscsiSet,

portal: 10.10.8.201,3260] successful.

[root@qaprod ~]# fdisk -l

Disk /dev/sda: 500.1 GB, 500107862016 bytes

255 heads, 63 sectors/track, 60801 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

Disk identifier: 0x00018542

Device Boot Start End Blocks Id System

/dev/sda1 * 1 64 512000 83 Linux

Partition 1 does not end on cylinder boundary.

/dev/sda2 64 60802 487873536 8e Linux LVM

Disk /dev/dm-0: 53.7 GB, 53687091200 bytes






Disk /dev/dm-0 doesn't contain a valid partition table







263










Disk /dev/sdb: 1073 MB, 1073741824 bytes





Disk identifier: 0x4408a443

Device Boot Start End Blocks Id System

/dev/sdb1 1 131 1045504 7 HPFS/NTFS

Partition 1 has different physical/logical endings:

phys=(129, 254, 63) logical=(130, 42, 40)

[root@qaprod ~]# fdisk /dev/sdb1

WARNING: DOS-compatible mode is deprecated. It's strongly recommended to

switch off the mode (command 'c') and change display units to

sectors (command 'u').

Command (m for help): m

Command action

a toggle a bootable flag

b edit bsd disklabel

c toggle the dos compatibility flag

d delete a partition

l list known partition types

m print this menu

n add a new partition

o create a new empty DOS partition table

p print the partition table

q quit without saving changes

s create a new empty Sun disklabel

t change a partition's system id

u change display/entry units

v verify the partition table

w write table to disk and exit

x extra functionality (experts only)

Command (m for help): n

Command action

e extended

p primary partition (1-4)

264


p

Selected partition 1

First cylinder (1-1008, default 1):

Using default value 1

Last cylinder, +cylinders or +size{K,M,G} (1-1008, default 1008):

Using default value 1008

Command (m for help): w

The partition table has been altered!

Calling ioctl() to re-read partition table.

</codedoc>

With CHAP authentication

For all targets

1. Open /etc/iscsi/iscsid.conf using an editor and then specify the details for the following section:

<codedoc code node.session.auth.authmethod = CHAP

node.session.auth.username = <CHAP username>node.session.auth.password = <CHAP secret>

node.session.auth.username_in= <Mutual CHAP username> (give only when you are doing mutual CHAP)

node.session.auth.password_in = <Mutual CHAP secret> (give only when you are doing mutual CHAP)

discovery.sendtargets.auth.authmethod = CHAP

discovery.sendtargets.auth.username = <CHAP username>

discovery.sendtargets.auth.password = <CHAP secret>

discovery.sendtargets.auth.username = <Mutual CHAP username> (give only when you are doing mutual CHAP)

discovery.sendtargets.auth.password_in = <Mutual CHAP secret> (give only when you are doing mutual CHAP)

</codedoc>

2. Restart the iSCSI daemon using the following command: /etc/init.d/iscsi restart.

3. Follow the procedures in the section "Without CHAP authentication".

For as specific target

Specify the details as follows at command prompt:

iscsiadm -m node -T "<Target_Name>" -p <VSM_IP_address>:<port> --

op=update --name node.session.auth.authmethod --value=CHAP

iscsiadm -m node -T "< Target_Name >“ -p < VSM_IP_address >:<port> --

op=update --name node.session.auth.username --value=<CHAP_username>

iscsiadm -m node -T "< Target_Name >“ -p < VSM_IP_address >:<port> --op=update

--name node.session.auth.password --value=<password>

iscsiadm -m node –T "<Target_ Name>" -p < VSM_IP_address >:<port> -l

265


Resizing iSCSI LUN

The following sections explain how to expand your iSCSI storage volume on CentOS:

l "Mapping iSCSI LUN"

l "Editing Quota Size (in ElastiCenter)"

l "LUN expansion and resizing file system (on CentOS Console)"

Exporting the iSCSI LUN (On CentOS Console)

In the CentOS console, run the commands provided in the table in sequence to export the iSCSI LUN:

Step Task Command Example Details

1 Discover the iSCSI tar-gets.

iscsiadm -m discovery -tst -p <Target IP Address>

iscsiadm -m discovery -t st-p 11.11.11.11

Outputs the completetarget name.

2 Log in to the iSCSI target. iscsiadm -m node -T <Com-plete Target Name> -l

iscsiadm -m node -Tiqn.2014-02.FCAc-c.Test:FCAcciscsivol1 -l

3 Display the current par-titions on the hard disk.

fdisk –l The partition is dis-played in the output.

4 Create a file system onthe hard disk drive.

mkfs.ext3 <Partition name> mkfs.ext3 /dev/sdc At prompt, confirm bytyping y as shown in thefollowing example:/dev/sdc is entiredevice, not just one par-tition! Proceed anyway?(y,n) y

5 Create a directory. mkdir <Directory name> mkdir /mnt/cbdata

6 Mount the partition intoyour directory.

mount <Partition name><Directory name>

mount /dev/sdc /mnt/cbdata-o sync

7 (Optional) Retrieve thecurrent size of the moun-ted LUN.

df –kh

8 Copy some data to thedirectory /mnt/cbdata.

cp

9 Unmount directory fromthe disk.

umount <Directory name> umount /mnt/cbdata

10 Log out of iSCSI target. iscsiadm -m node -u -T<Complete target name> -p<Target IP Address>

iscsiadm -m node -u -Tiqn.2014-02.Ac-c1.TSM3:TSM3TSM3ISCSI11 -p11.11.11.11

Editing Quota Size (in ElastiCenter)1. In the ElastiCenter, select Storage Volumes.

2. In the Storage Volumes page, click Settings > Edit.

266


3. Increase the Quota size of the LUN and then click Save.

LUN expansion and resizing file system (on CentOS Console)

In the CentOS console, run the commands provided in the table in sequence to expand the LUN and resize the file system:

Step Task Command Example Details

1 Log in to iSCSI target. iscsiadm -m node -T <Com-plete target name> -p<Target IP Address> –l

iscsiadm -m node -Tiqn.2014-02.Ac-c1.TSM3:TSM3TSM3ISCSI11 -p16.10.10.3 –l

2 Display current partitions onthe hard disk.

fdisk –l Check the outputto see if the disksize hasincreased.

3 Mount the partition into thedirectory.

mount partition name <Dir-ectory name> -o sync

mount /dev/sdb /mnt/cbdata/-o sync

4 Resize the expanded volume. resize2fs /dev/sdb/ totalspace of volume

resize2fs /dev/sdb/ 5G In this example,5G is the totalspace on the res-ized Volume.

5 Retrieve the current size ofthe resized LUN.

df –kh

OnWindows Server 2008

Discovering the iSCSI target


267


1. Ensure that you have enabled iSCSI. For details, see "Data access using iSCSI".

2. OnWindows Server 2008, launch iSCSI Initiator (Programs > Administrative Tools > ISCSI Initiator).

3. In the Target tab, specify the target (tenant IP address).

4. Click Quick Connect. TheQuick Connect screen appears, displaying the target name.

5. Click Done.

Create storage drive to use the iSCSI target


268


2. Right-click the unallocated space (the highlighted section) and then select Initialize Disk. The Initialize Disk dialog boxappears.

269


3. Click OK.

4. Right-click in the space and then select New Simple Volume as shown in the following screen:

270


The New Simple Volumewizard appears.

5. Follow the instructions to complete the wizard, to create the partition.

6. Ensure that the created disk is added to the Disk management console.

On Kernel-based Virtual Machine

Note:The following section provides procedures performed on RHEL 6.3. The procedures might be similar on other flavors ofLinux.

1. On your Linux desktop, select Applications > System Tools > Virtual MachineManager.

2. Select a host machine.

3. Click Edit > Connection Details.

271


4. Click Storage tab.

5. Click the Add Storage Pool + icon at the bottom left side of the dialog box. The Add a New Storage Pool wizard appears.

6. Specify a name for the Pool, select iSCSI as the type, and then click Forward.

272


7. In the following screen, specify the required details.

Field Description

Target Path Path of the target directory (to use for the Storage Pool).

273


Host Name Host name or IP address of the iSCSI target.

Source Path iSCSI target IQN.

IQN Select the check box and then enter the IQN for the initiator.

8. Click Finish.

Mapping CIFS share to the clients

OnWindows Server 2008/Windows Server 2012

The following procedures explain how tomapCIFS share from CloudByte ElastiStor onWindows Server 2008/Windows Server2012.



1. Ensure that you have enabled CIFS. For details, see "Data access using CIFS".

2. OnWindows Server, launchMapNetwork Drive dialog box (Computer > MapNetwork Drive).

3. Specify IP address of VSM and its access path in the following format:\\<VSM IP Address>\<Access Path>.

4. Click Finish.

5. (If you are using the VSM for the first time) At prompt, specify the credentials in the Enter Network Password dialog box. Thisis the username and password that you have created for the account user in ElastiCenter.

274


6. Click OK.

7. (If again prompted) Specify the credentials in the Enter Network Password dialog box.

8. Ensure that CIFS share is connected, as shown in the following screen:


The following procedures explain how tomapCIFS share from CloudByte ElastiStor onWindows Server 2012.

Discovering the CIFS share


275


1. Ensure that you have enabled CIFS. For details, see "Data access using CIFS".

2. OnWindows Server 2012, launchMapNetwork Drive dialog box (Computer/Shortcut on Desktop to Computer > MapNet-work Drive).

TheMapNetwork dialog box appears.

3. (If you are using the VSM for the first time) At prompt, specify the credentials in the Enter Network Password dialog box. Thisis the username and password that you have created for the account user in ElastiCenter.

276


4. Click Finish.

5. (If again prompted) Specify the credentials in the Enter Network Password dialog box.

6. Ensure that CIFS share is connected, as shown in the following screen:

Mapping NFS share to clients

On Red Hat Enterprise Linux

Note:Themapping procedures are similar for other flavors of Linux.

1. Ensure that you have enabled the NFS. For details, see "Data access using NFS".

2. On your RedHat Enterprise Linux server, at command prompt, specify the following command to verify the NFS export:showmount -e <tenant IP address>

277


If correctly exported, the path to the filesystem along with the authorized network is displayed.

3. Create a directory to mount the NFS using the following command: mkdir <directoryname>

4. Mount the NFS on this directory using the following command: mount -o mountproto=tcp <tenant IPaddress:>/<exported path><directoryname>

5. Enter the following command: mount

6. If successfully mounted, the following is displayed in themount list: <tenant IP address>:<exportpath> on <dir-ectoryname> type nfs (rw,addr=<tenant IP address>)

Sample CLI snapshot

[root@localhost ~]# showmount -e 20.10.26.199Export list for 20.10.26.199:/acctsample (everyone)

#Here everyone implies that the exported path is shared to all (as configured in NFS page).If you specify a specific client IP, that IP is displayed.

[root@qaprod ~]# mkdir /mnt/nfslocal[root@qaprod ~]# mount -t nfs -o vers=3,proto=tcp,nolock,sync 20.10.26.199:/acctsample /mnt/nfslocal/[root@qaprod ~]# mount

/dev/mapper/VolGroup-lv_root on / type ext4 (rw)proc on /proc type proc (rw)sysfs on /sys type sysfs (rw)devpts on /dev/pts type devpts (rw,gid=5,mode=620)tmpfs on /dev/shm type tmpfs (rw,rootcontext="system_u:object_r:tmpfs_t:s0")/dev/sda1 on /boot type ext4 (rw)none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)gvfs-fuse-daemon on /root/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev)20.10.26.199:/acctsample on /mnt/nfslocal type nfs(rw,sync,vers=3,mountproto=tcp,addr=20.10.26.199,mountaddr=20.10.26.199)

On Kernel-based Virtual Machine

Note:The following section provides procedures performed on RHEL 6.3. The procedures might be similar on other flavors ofLinux.

1. On your Linux desktop, select Applications > System Tools > Virtual MachineManager.

2. Select a host machine.

3. Click Edit > Connection Details.

278


4. Click Storage tab.

5. Click the Add Storage Pool + icon at the bottom left side of the dialog box. The Add a New Storage Pool wizard appears.

6. Specify a name for the Pool, select Network Exported Directory as the type, and then click Forward.

279


7. In the following page, specify the required details.

Field Description

Target Path Path of the target directory (to use for the Storage Pool).

Format Set the Format option to NFS.

Host Name Host name or IP address of the NFS Server.

Source Path Enter the NFS path (path on the Host that is being shared).

8. Click Finish.

280

CHAPTER XII Frequently asked questions


Product FAQs

Is it possible to scale-out using CloudByte ElastiStor?

Each CloudByte ElastiStor node can be scaled to 1 PB of raw storage capacity. You can linearly increase capacity and performanceby adding additional ElastiStor nodes.

Can I use ElastiStor to improve the performance of storage solutions such as EMC Symmetrix orHitachi Unified Storage?

Like other storage systems, CloudByte ElastiStor is an independent storage system with distinct architecture.

ElastiStor does not improve performance but shares the performance between competing applications granularly in a desired way.However, CloudByte recommends you to use raw disk for storagemedia rather than storage from other vendors.

Note: In ElastiStor 1.4, support is limited to SAS connected disk arrays (and not any other storage product beneath).

How many flavors and versions of CloudByte ElastiStor are available in the market?

CloudByte ElastiStor 1.0 was the pioneering release of the product. The ongoing release is CloudByte ElastiStor 1.3.2.

Which RAID levels are supported by CloudByte ElastiStor?

RAID5, RAID6, RAID10, RAID50, RAID60.

Are there any CloudByte ElastiStor management plug-ins available?

Yes. The plugins for VMWare vCenter andOpenStack Cinder are available off-the-shelf.

Can I use the same IP address for multiple tenants?

Each tenant is identified by unique IP address.

Does ElastiStor support iPV6?

Current version of CloudByte ElastiStor does not have built in support for iPV6.

What is the skill set required to manage and configure ElastiStor?

Basic storage administration (SAN and NAS) and networking skills.

Provide ElastiStor specifications in terms of the supported number of logical table entries, forexample, the maximum number of simultaneous tenant networks, maximum number of storageclients, maximum number of LUNs, maximum number of NFS mount points, etc.

ElastiStor does not have theoretical limitations on the number of Storage Volumes, VLANs and VSMs per Node as revealed in thetests done with 12000 Volumes. Also, ElastiStor supports unlimited snapshots and clones. For details, see ElastiStor Scalability

281

http://www.docs.cloudbyte.com/elastistor-scalability-results/


The CloudByte documentation (http://www.docs.cloudbyte.com/enabling-out-of-band-communication-between-nodes-in-an-ha-group/ ) refers to configuring IPMI between Nodes in an HA Group. Is such aconfiguration required for ElastiCenter HA also?

No. IPMI based out of band communication between peer Nodes is used during Node HA process when necessary. This con-figuration is not mandatory for ElastiCenter HA. However, it is helpful to have the IPMI configured for the regular remote access andtroubleshooting.

Is SNMP monitoring support available on ElastiStor?

Yes.

Is SNMP based IOPS and capacity monitoring support available on ElastiStor?

SNMP support for volume statistics such as IOPS will be available in the future releases.

How to configure a spare disk?

Follow the procedures at the URL http://www.docs.cloudbyte.com/disk-replacement-scenarios/.

In pages where Graphs ought to appear, I see a message invalid data throughout ElastiCenter.Why?

Try using a different browser.Adobe Flash is required to see the charts. For details, see http://www.docs.cloudbyte.com/wp-content/uploads/CloudByte_troube-shooting_invalid-data_graphs.pdf. Future releases will provide HTML5 support.

How to set up a secondary ElastiCenter manually?

This scenario applies if you haven't configured ElastiCenter HA during installation. If ElastiCenter fails, you can promote the sec-ondary ElastiCenter to primary ElastiCenter. For procedures, see Setting up secondary ElastiCenter

I am unable to access ElastiCenter on the browser Chrome (version 45). Is there a workaroundto access ElastiCenter?

Yes. Do the following to access ElastiCenter onGoogle Chrome version 45:

1. SSH to ElastiCenter as root.

2. Download the server.xml file from the following URL: http://cloudbyte.com/downloads/server.xml

3. Backup the existing xml file in the location /usr/local/apache-tomcat-6.0.33/conf/server.xml

4. Run the following command to replace the existing server.xml file with the new xml (that you downloaded): cp new_server.xml /usr/local/apache-tomcat-6.0.33/conf/server.xml

5. Run the commands to restart Tomcat service:

1. /usr/local/apache-tomcat-6.0.33/bin/shutdown.sh

2. pgrep java

3. (Optional, if the java service is running) pkill java

4. /usr/local/apache-tomcat-6.0.33/bin/startup.sh

6. Open a new tab in the browser.

7. Launch ElastiCenter.

282

http://www.docs.cloudbyte.com/enabling-out-of-band-communication-between-nodes-in-an-ha-group/

http://www.docs.cloudbyte.com/disk-replacement-scenarios/

http://www.docs.cloudbyte.com/wp-content/uploads/CloudByte_troubeshooting_invalid-data_graphs.pdf

http://www.docs.cloudbyte.com/cloudbyte-elastistor-product-faqs/promoting-secondary-elasticenter-and-backing-up-configurations/

http://cloudbyte.com/downloads/server.xml


I encountered the error weak ephemeral Diffie-Hellman key while trying to access ElastiCenter. Iam using Mozilla Firefox version 39 browser. Is there a workaround to access ElastiCenter?

Yes. Do the following to access ElastiCenter onMozilla Firefox version 39:

1. LaunchMozilla Firefox browser.

2. In the Address bar, type the following:

about:config.

3. In the Search bar, search for the following parameter:

security.ssl3.dhe_rsa_aes_128_sha

4. Set the parameter to

false (security.ssl3.dhe_rsa_aes_128_sha = false)by doing either of the following:

l Right click the parameter and then select Toggle.

l Double click the parameter.

5. Repeat step 3 and step 4 for the parameter

security.ssl3.dhe_rsa_aes_256_sha

6. Open a new tab in the browser.

7. Launch ElastiCenter.

I cannot access the configuration file config.xml from ElastiCenter. How can I access it oth-erwise?

The following procedures let youmanually restore a earlier file version to its original location:

1. Move the Node toMaintenancemode. For details, see "Configuring High Availability"

2. Run the following command to copy the latest config.xml file from the backup location: cp /var/-backups/node/config.xml YYYY-MM-DD-HH-00-00 /cf/conf/config.xml.

YYYY-MM-DD-HH-00-00 is the date and time when the file was backed up .

3. Move the Node to available mode. For details, see "Configuring High Availability"

283


Does ElastiStor support NPIV?

No

What is API key?

The API Key is used to authenticate the CloudByte REST API token requests. See, "CloudByte ElastiStor API references" toobtain a Rest API Key.

Installation FAQs

I cannot access CloudByte ElastiCenter after successful installation. Why?

This might be because of one of the following reasons:

l Java Service is not running

l Database is not up and running

To resolve this issue, do either or both of the following:

1. In the ElastiCenter server, at command prompt, run the following command: pgrep mysql.

If you do not get theMySql process ID, run the following command to bring up the database service: service mysql-server onerestart

2. In the ElastiCenter server, at command prompt, run the following: pgrep java.

If you don't get the Java process ID, run the following command to bring up the Java service: service cbdevman onere-start

3. Try accessing ElastiCenter.

Can you explain the memory requirements for installing CloudByte ElastiStor?

The following table provides the details. The values are same for installing both Node and ElastiCenter. Also, the same values applyif you install both Node and ElastiCenter on the samemachine:

Memory Minimum Recommended

RAM 1 GB 8 GB or higher

Hard Disk 20 GB (2.4 X RAM size + 32 GB) or higher. For example, if the RAM is 8 GB, for better performance, 52 GB orhigher is recommended

If you do not comply with the recommended hard disk memory, youmight encounter amessage that prompts correction (for betterperformance). However, if youmeet theminimummemory requirements, even if you ignore themessage, installation will be suc-cessful.

284


I get the message ' Installation cannot proceed as no NIC is present' . What will I do?

You encounter this issue because the NIC that you use is not supported. For a list of supported NICs and other hardware, see theHardware Compatibility List.

I get the message ' Installation cannot proceed as no disk is present' . What will I do?

This is because the internal RAID/SAS card in the server is not supported. For a list of supported disks and other hardware, see theHardware Compatibility List.

What if I get a mountroot during installation?

Retry installing by reinserting the CD. Also ensure that there are no network issues.

If the issue persists, use the following command to proceed: mountroot > cd9669:/dev/cd0 ro

Why am I not able to access the system after the first reboot post installation?

Verify the following:

n ElastiStor has correctly rebooted. This is indicated by the root login prompt when the installation concludes.

n Network port connectivity

n Network details are correctly added

Will the installation rollback on failure?

Yes

Explain the upgrade options for ElastiStor?

You can upgrade based on the option that you have selected in the existing installation. That is, if you have opted to install bothNode and ElastiCenter, then you have the option to upgrade both.

What happens to the installation when there is a hard disk failure?

Installationmay or may not be successful, but the product is unlikely to function.

What happens if the CD ROM gets disconnected while installing?

Reconnect and start the installation from the beginning.

Do I need an internet connection for installing CloudByte?

No. But it is recommended because it facilitates time synchronization with NTP.

What happens if the network connection is lost and system boots after installing?

It waits for the connectivity to be restored and then boots up.

What happens to the installation during a power loss?

You have to reinstall.

285




Can I install ElastiStor on my laptop?

Yes. If youmeet theminimum hardware requirements, the installation is not device dependent.

Can I install ElastiCenter on a virtual machine?

Yes. CloudByte recommends that you have aminimum RAM of 8 GB on your VM.

Is ElastiStor installation possible over SAN boot?

Yes

Can I downgrade my ElastiStor installation?

Not currently supported

Can I do a network installation of CloudByte ElastiStor?

To perform a network installation of CloudByte ElastiStor, you have to set up a PXE server and then perform a network boot.

After I installed ElastiCenter on ESX 5.0, my virtual machine stopped responding. Why?

If your virtual machine on ESX 5.0 hangs, do the following:

1. (In the ElastiCenter command prompt) Run the following command: sysctl kern.timecounter.hardware=ACPI-fast

2. Run the following command to edit the loader.conf file: vi /boot/loader.conf

3. Append the loader.conf with the following line: kern.timecounter.hardware=ACPI-fast

I did not come across the ElastiCenter HA option while installing ElastiCenter. Why?

The option to configure ElastiCenter HA is available only when you choose the installation option Both ElastiCenter and Node i.e.,installing both ElastiCenter and Node on a single machine. For details, see http://www.docs.cloudbyte.com/installing-both-elasti-center-and-node/

Can I back up ElastiCenter even if I haven’t configured ElastiCenter HA?

Yes. You can setup a secondary ElastiCenter (secondary setup) to ensure backup in the case your primary ElastiCenter (primarysetup) fails. If a failure occurs, you can promote your secondary setup. For details, see http://www.docs.cloudbyte.com/cloudbyte-elastistor-product-faqs/promoting-secondary-elasticenter-and-backing-up-configurations/

Does installing ElastiCenter in HA mode involve any manual procedures?

No. During installation, one ElastiCenter is installed as primary and the second one as secondary.

Is it mandatory to have ElastiCenters and ElastiStor Nodes on the same subnet?

Not necessary for them to be on the same Subnet as long as the IP address is reachable.

286



Pool FAQs

What is the recommended storage pool RAID configuration?

If you need high performance, mirroring is recommended. For larger SATA drives (for example 4TB), CloudByte recommendsRAIDz2.

How can I expand the tpool size?

1. SSH to the Node.

2. Run the following command: cd /usr/local/cb/bin.

3. Run the following to expand the tpool size: ./expandtpool OS_disk_name. For example, ./expandtpool ada0

The size of the tpool increases by 10G.

Can I create a mixed disk pool, for example with SATA and SSD disks?

Yes. But CloudByte does not recommend this because you get only the performance of the lowest disk; in this case SATA.

How many log devices can I attach to a Pool?

Depends on the workload. For instance, in aWRITE intense scenario, a higher ZIL improves yourWRITE performance.

How much L2ARC should I add to the Pool?

Depends on the workload. For instance, in a READ intense scenario, a higher L2R improves READ performance.

How to improve the performance of L2ARC and the log device?

L2ARC and log device are intended to improve the READ andWRITE respectively. Therefore, high performance SSD storage isrecommended for an L2ARC or ZIL

What Pool do you recommend for a data protection use case?

A large RAIDz Group with high performance data disks and one or two parity disks are ideal.

What is the recommended RAID-Z configuration?

For RAID-Z1, RAID-Z2, and RAID-Z3 CloudByte recommends a 2^n + p configuration.

Parameter Description

n RAID-Z 1, RAID-Z 2, and RAID-Z 3.

p Parity: p=1 for RAID-Z1, p=2 for RAID-Z2 and p=3 for RAID-Z3 where RAID-Z1 = 3, 5, 9, 17, …, RAID-Z2 = 4, 6, 10,18, …, RAID-Z3 = 5, 7, 11, 19, …

Is s ingle parity RAID equivalent to well known RAID 5?

Almost similar. Also it eliminates theWRITE HOLE problem in RAID 5.

Is double parity RAID equivalent to well known RAID 6?

Almost similar. Also, it eliminates theWRITE HOLE problem in RAID 5.

287


What is RAID triple parity and when should I use it?

RAID triple parity tolerates failure of three disks in a RAID group. Use it when you handle critical data and want to evade chances ofdisk failure.

Do you recommend NFS/CIFS/ISCSI/FC setup on the same Pool or should I have the access pro-tocols set up on different Pools?

It doesn't matter. Configure based on the usage and requirement.

Can a VSM span multiple Pools?

Not in the current release.

Is Pool IOPS derived from the disk?

Disk plays a role in deriving IOPS, but is also based on RAM and L2ARC.

Can I extend the Pool IOPS by extending the Pool?

Yes

Can I delete a Pool with active Storage Volume?

No

Can I identify the failed disks in a Pool?

ElastiCenter Alerts provides you with the details.

What happens if you pull a disk from a Pool that has a spare disk?

Spare disk is attached to the Pool.

What happens if you pull a disk from a Pool that does not have a spare disk?

Leaves the Pool in a degraded list.

How do I disable SCSI reservation on disks?

Warning: Do not disable SCSI reservation on production environments. The following procedures are recommended only fortroubleshooting on development and testing setups.


2. Run the following command touch /etc/disablescsi.

When youmove your setup from test to production environment, ensure that all disks are SCSI reservation-enabled.

Do the following to enable SCSI reservation:


2. Run the following command rm /etc/disablescsi.

288


How will I clear the pools created in an older setup?

If you add a new Pool on the disks being consumed by existing Pool, it will be deleted automatically. However, to manually clear thePools from an older setup, do either of the following (depending on the scenario):

l Reinstalled ElastiStor without removing the Pools: In the Nodes page, go to the section Existing Pools (at thebottom of the page) and then click the Clear link under the columnClear.

l Connected disk arrays (with Pools) to a different Node: Apply Refresh Hardware after connecting to the newNode. All existing Pools are displayed in the Nodes page. Go to the section Existing Pools (at the bottom of the page) andthen click the Clear link under columnClear.

How do I configure partial failover?

CloudByte ElastiStor supports selective failover of Pools. If partial failover is enabled, in the case of network failure, the secondarynode in the cluster takes over.

By default, partial failover is enabled. If not enabled, go to the Pool settings and enable.

1. In ElastiCenter, select Pools in the left pane.

2. In the Pools page, select the Pool for which you want to enable partial failover.

3. In the Actions pane, select Settings.

4. Click Edit and then enable partial failover.

Access Protocol FAQs

Can I use the same secret key for mutual authentication in the case of LUN export?

No. The key must be unique.

What is the maximum number of NFS or iSCSI LUNs that you can have on ElastiStor?

You can create N number of NFS or iSCSI LUNs, based on your requirement.

289


Can I send an I/O of 1MB size in a single request on NFS?

Yes. NFS volumes are currently allowed to bemounted with an rsize/wsize of 1M.

Can I disable the configuration options related to the protocols I do not use, in ElastiCenter?

Yes. The current release supports disabling of FC-specific options.

1. In the ElastiCenter, click admin > Global Settings.

2. Go to ui.protocol.enablefc, click the Edit icon, and then specify false.

3. For the changes to take effect, either relogin to ElastiCenter or refresh the browser.

All FC-specific options disappear from ElastiCenter UI.

When I use FC I see a LUN with the ID 0 and size 1MB displayed though i did not create it. Why?

This is a system LUN and is not for use.

Deleting data on my CIFS share doesn’t increase the free space. Why?

This is because the data is not permanently deleted. Instead, data is moved to the CIFS share recycle bin. To delete data per-manently, do the following:

On ElastiCenter

1. In the Storage Volumes page, select a CIFS Storage Volume.

2. Click Manage CIFS configuration in the actions pane.

3. Click Edit and then disable the CIFS share Recycle bin option.

290


4. Click Save.

OnWindows client

If you have not disabled this option (earlier) and then data is already moved to the CIFS share Recycle Bin, to clear the data, youmust do the following:

291


l (If Show hidden files and folders option is enabled) Go to the folder .recycle and then clear the data.

l (If Show hidden files and folders option is disabled) Do the following:

1. Go to Control Panel > Folder Options > View tab.

2. Enable hidden files, folders, and drives. Now you can view the .recycle folder.

3. Clear the data.

I am on a common NFS Share. Is there a way to restrict an other user on this share from deletingmy files?

Yes. If you want a home directory kind of nfs share, when configuring the NFS Storage Volume (see the following screen), set theoption Map as Root Users to No.

I changed the primary IP address of the VSM. Now I cannot access the iSCSI Storage Volumes.What should I do?

1. Log out of all the iSCSI Volumes on the clients.

2. Delete the login entries of the VSM on the clients.

3. Relogin to the clients with the new IP address.

292


Storage Volume FAQs

Can I migrate a Storage Volume?

No. You canmigrate only VSMs.

Can I mirror a Storage Volume?

You can do a DR of your storage volume.

Can I keep Storage Volume in a sleep state?

No.

Can a Storage Volume span multiple pools?

No.

Will enabling compression and deduplication impact performance?

Yes, they can significantly impact. Compression and deduplication are CPU andmemory intensive processes.

What is the right latency for Storage Volume?

Depends on the disk from which the Pool is created. For example, SSDs are of lower latency when compared to SATA.

293


I enabled READ-only property for iSCSI Volume mounted on a Windows client, and observe thatthe property hasn't taken effect. How can I make the Volume READ only?

For the READ-only property to take effect, disconnect the iSCSI Volume and then remount it on the Client.

Can I have Storage Volumes with same names under different accounts?

Yes.

Can the storage capacity be changed after configuring a Storage Volume?

Yes.

QoS FAQs

What is QoS?

Assuring your client/application the promised IOPS, throughput, and latency.

Is QoS applicable only to Cloud Service Providers?

No. QoS is applicable to enterprise scenarios as well.

What is the maximum IOPS I can set?

Themaximum IOPS directly depends on the underlying storage. SSD is the best choice followed by SAS and SATA.

Is QoS disk dependent?

Yes

Is QoS a licensed option?

No

What is the difference between dedicated and shared options?

Dedicated: If the Storage Volume needs exclusive QoS configuration.

Shared: If the Storage Volume can share the QoS configuration with another Storage Volume.

The following figure explains the concepts:

294


For example, if both Storage Volume 1 and Storage Volume 2 are committed to busy service round the clock, Dedicated would beyour choice. Whereas, if Storage Volume 1 and Storage Volume 2 represent services in a cross-geographical setup where activitiesrelated to Storage Volume 1 end toward the time activities related to Storage Volume 2 start, Shared would be your choice.

Do I have to set all the QoS parameters (IOPS, Latency, and Throughput)?

No. IOPS is amust. Depending on requirement, you can choose to configure Latency and Throughput.

What is the difference between IOPS, Latency, and Throughput?

IOPS: The READ/WRITE capability.

Latency: Time that ElastiStor takes to acknowledge a READ/WRITE back to the client.

Throughput: Data that can be transferred as part of READ/WRITE operations.

Is it a must to set QoS?

Yes

Can I increase/decrease QoS on-the-fly?

Yes

Is QoS guaranteed in an HA environment?

Yes

What are the recommended QoS settings in an HA environment?

In multiway HA, for example in a two-way HA, each Node should be provisioned amaximum of up to 50 percent of the capacity

295


Can I set QoS for my backups?

Yes

Can I set QoS for my DR operations?

Yes

Can I get the same QoS on the secondary Node in the case of a disaster?

Yes, provided you have the same or better infrastructure quality.

What is the upper limit for setting FC LUNS?

255 per target

How do you calculate IOPS, latency, and throughput?

Throughput = IOPS * Block Size.

Latency depends on the following factors:

l Workload characteristics such as L2ARC Size and ZIL size.

l Seek time on the disk

l Workload access patterns such as sequential, random, read only, read write, write only.

l Speed of the underlying disks, that is, if spinning disks such as SAS or SATA or faster ones such as SSDs

l Dataset cached in the L1ARC caches and L2ARC caches.

l Presence of Intent log ZIL to log the writes and respond quickly.

l How the ZIL is configured, that is if it is on spinning disks or SSD disks.

l Underlying RAID configuration, for example striped or mirrored.

l System load in terms of CPU cycles available.

Can I set QoS for my backups?

Not for the current release.

Can I set QoS for my DR operations?

Not for the current release.

Can I get the same QoS on the secondary Node in the case of a disaster?

Yes, provided the secondary node has same system configuration as the primary.

When do I use Grace?

Particularly useful when you check if the IOPS and throughput set for a particular Storage Volume are correct. If a particular StorageVolume is continuously eating up grace, it is an indication that it is over-provisioned and deserves more IOPS/throughput.

296


How can I compare the QoS values of selected Volumes?

1. In CloudByte ElastiCenter, click Pools in the left pane.

2. In the Pools page, select the Pool for which you want to compare the QoS values of the selected Storage Volumes.

3. Select the Storage Volumes.

4. Select the QoS parameter (IOPS, throughput, or latency) that you want to compare.

5. Click Update.

The chart on top of the page displays the comparison with time stamp as the reference.

High Availability FAQs

The Node fails to move back to available mode. Why?

In the case of failure, verify the storage connectivity and check for any error messages in the View Jobs (HA Group > select the spe-cific HA Group > actions pane > Tasks). If you cannot resolve, contact Technical Support.

How do I configure redundant management NIC?

You can establish a highly available network connection with a redundant (backup) network interface. The redundant network inter-face protects against network failures. If the activemanagement network interface becomes unavailable, the system automaticallyswitches to the redundant interface tomaintain availability.

Assumptions

l Management interface is named em0

l Backup interface is named em1

l Node has been added on ElastiCenter. For details, see Adding a Node.

Procedures

Do the following to set up em1 as the redundant network interface:

1. Log in to ElastiStor Node as root.

2. Run the following command to edit the config.xml: vi /cf/conf/config.xml

3. In the config.xml, search for the tag <type>webgui</type> and specify the following as the next line: backupi-f>Management_Interface/backupif>For example,

<if>em0</if><type>webgui</type><backupif>em0</backupif>

4. In the config.xml, search for <if>Backup_interface</if> and specify the following as the next line:<backupi-f>Backup_interface</backupif>For example,

<if>em1</if><backupif>em0</backupif>

Sample Snippet

<lan><uuid>fbfaf211-fd6d-4a70-984a-9642720b9d61</uuid><if>em0</if><backupif>em1</backupif><ipaddr>192.168.16.23</ipaddr><subnet>24</subnet>

297

http://www.docs.cloudbyte.com/what-is-elasticenter/adding-a-host/


<type>webgui</type><gateway>192.168.16.1</gateway><status>active</status><mtu>1500</mtu></lan><lan><status>active</status><subnet>none</subnet><uuid>5b0fa24e-463c-4933-817b-94b94129445c</uuid><ipaddr>none</ipaddr><mtu>1500</mtu><type>none</type><gateway>none</gateway><if>em1</if><backupif>em0</backupif>

</lan>

How will I protect my setup against network failures?

See the FAQHow do I configure redundant management NIC?

What should I do if my primary ElastiCenter (in ElastiCenter HA setup) goes down?

For troubleshooting this issue, see Troubleshooting ElastiCenter HA.

ElastiCenter displays the message HA ring broken. What does the message refer to?

The HA ring brokenmessage is displayed when themanagement network fails. Youmay have to analyze the logs to find outthe reason for the network failure.

NIC FAQs

What do the terms degraded, active, and inactive imply?

Term Description

Degraded The interface is active on one Node but inactive on another in the same HA Group.

Active The interface is connected to the network on all Nodes in the HA Group.

Inactive Either the interface is not connected to the network or it is connected but not active (possiblybecause the interface status could not be fetched while adding the Node to the HA Group. Inthis case, in the Node page, click Network Interface Cards and then click Refresh).

Global Settings

Global settings let the administrator configure various options used to control CloudByte ElastiStor. For example, email con-figuration. To configure global settings,


2. Select a target category. A list of settings is provided by default.

3. Edit the relevant settings.

Note: You cannot add new settings.

298

http://www.docs.cloudbyte.com/troubleshooting-elasticenter-ha/


This section lists the global settings options available on the ElastiCenter.


alert.email.addresses Comma separated list of email addresses usedfor sending alerts.

Email ID you specified in the UpdateProfile section in ElastiCenter.

alert.email.enable If true, emails are sent to the specified emailaddresses.

TRUE

alert.email.sender Sender of the alert email.

alert.email.severity Set the email severity level to one of the fol-lowing:

l 1 - ERROR

l 2 - CRITICAL

l 3 - WARNING

l 4 - INFO

4

alert.smtp.host SMTP hostname to send alert emails. localhost

alert.smtp.password Password for SMTP authentication (applies onlyif alert.smtp.useAuth is true).

alert.smtp.port The port of the SMTP server.587

alert.smtp.useAuth If true, SMTP authentication is enabled to sendemails.

False

alert.smtp.username Username for SMTP authentication (applies onlyif alert.smtp.useAuth is true).

autoSupport.config.details Send configuration details to CloudByte as partof an auto support email.

Yes

autosupport.httpnotify.apikey API key of CloudByte auto support server.

autosupport.httpnotify.enable Send HTTP post request to CloudByte auto sup-port server. False

autosupport.httpnotify.ipaddress The IP address of CloudByte auto support server.

autoSup-port.infastructure.statistics

Send infrastructure statistics to CloudByte aspart of an auto support email.

Yes

autoSupport.product.alerts Send product related alerts to CloudByte as partof an auto support email.

Yes

autoSupport.reg.details Send registration details to CloudByte as part ofan auto support email. Yes

autoSupport.tenant.statistics Send tenant statistics to CloudByte as part of anauto support email.

Yes

bad.disk.threshold The number of defects grown on disk that willdeclare it as a bad disk

512

299



cache.multiplication.factor The number of Cache READs which equal a diskREAD. The default value is 3 which means 1 diskREAD equals 3 cache READs. Only a valuebetween 2 and 6 is supported.

3

devman.backup.files.retain The number of backup files to be retained. 5

devman.backup.path The path of ElastiCenter backup directory. /var/devman/backups

devman.backup.scheduling.interval The interval time for ElastiCenter backup. 1

devman.build.number The build version of ElastiStor. Available in /cbdir/buildversion

monitor.purge.retain.days The number of days that the data has to beretained for.

60

monitor.qos.stats.interval The time interval (in seconds) taken for QoS stat-istics to be retrieved from Nodes.

300

monitor.refresh.interval The time interval (in seconds) required torefresh the Storage Volumes page.

300

monitor.storage.stats.interval The time interval (in seconds) taken for storagestatistics to be retrieved from Nodes.

300

monitor.system.purge.retain.days The number of days that the data has to beretained for.

7

no.of.nodes.per.hagroup Limiting number of Nodes in an HA Group.

project.invite.required Specify if invitation confirmation is requiredwhen add account to project.

False

provisioning.sb.algo Algorithm used for auto-provisioning StorageVolume.

Best Fit

provisioning.sb.auto Enable/Disable auto-provisioning of StorageVolumes.

No

provisioning.tsm.algo Algorithm used for auto-provisioning VSM. Best Fit

provisioning.tsm.auto Enable/Disable auto-provisioning of VSMs. No

qos.penalty.allowed.delta Percentage (as a value between 1 and 100) up towhich penalty will be not be considered

10

qos.penality.blocksize Standard blocksize (in KB) per IOPS, used in pro-visioning and penality enforcement.

4

qos.penalty.blocksize.factor Considers block size for penalty calculations. Spe-cify yes to Enable and No to Disable.

No

qos.penalty.cachehit.factor Considers cache hit for penalty caclulations. Spe-cify yes to Enable and No to Disable.

No

qos.pen-alty.enforcement.timeinterval

The time interval (in seconds) for which penaltyhas to be applied on QoS values.

300

qos.penalty.readwrite.factor Considers READs vs WRITEs for penalty caclu-lations. Specify yes to Enable and No to Disable.

No

300



read.multiplication.factor Specify howmany READs equal a WRITE. Thedefault value is 4 which means 1 WRITE equals 4READs. Only a value between 3 and 6 is sup-ported.

4

report.email.sender Sender of report email.

report.smtp.host SMTP hostname used for sending out reportemails.

localhost

report.smtp.password Password for SMTP authentication (applies onlyif report.smtp.useAuth is true).

report.smtp.port The port of the SMTP server. 587

report.smtp.useAuth If true, SMTP authentication is enabled to sendemails.

False

report.smtp.username Username for SMTP authentication (applies onlyif report.smtp.useAuth is true).

ui.protocol.enablefc Enable/Disable all FC protocol related optionsfrom ElastiCenter.

True

ui.protocol.enablehaadmincreation Enable/Disable haAdmin creation options fromElastiCenter.

False

ui.protocol.enablezilmirroring Enable/Disable ZIL mirroring creation optionsfrom ElastiCenter.

False

provisioning.uncontrolled.iops Enable/Disable provisioning with uncontrolled(zero) IOPS. Specify TRUE to enable.

False

alert.qos.throttling.percentage Sends an alert if the throttled IOPS percentageexceeds the specified value and if aler-t.qos.throttle.count is encountered for thespecified number of times.

20

alert.qos.throttle.count Sends an alert if the number of consistent throt-tling exceeds the specified value and aler-t.qos.throttling.percentage exceeds thespecified value.

10

devman.enable.ipspaces Enable/Disable the Assigning real IPspaces feature.

False

CloudByte ElastiStor best practices

CloudByte recommends the following best practices tomake CloudByte ElastiStor work effectively:

Best practice Description Reference

Creating Pools Create Pools with same vendor/RPMs/sectors/cylinders Pool FAQs

Client Multipathing On client-side, ensure multipath-enabled ISCSI/FC LUN MPIODevice mapper

LAGG To avail HA, set up network LAGG on the Node Configuring interfaces

Network switch stacking Stack network switch to handle network failure

301

http://www.docs.cloudbyte.com/cloudbyte-elastistor-product-faqs/pool-faqs/

http://www.docs.cloudbyte.com/what-is-elasticenter/provisioning-a-tsm/

http://www.docs.cloudbyte.com/knowledge-base-articles/configuring-device-mapper-multipath-on-linux/



SAS Multipathing Set up SAS multipath for effective path failover SAS multipathing

Performance and redundancy Configure ZIL, cache, and hot spare drives for Pools for betterperformance and redundancy

Network MTUs Configure Network MTUs based on client application recom-mendation

MTU revision Revise MTU prior to starting I/O on client. Changing the MTUlater might disrupt I/O

Secondary ElastiCenter Set up secondary ElastiCenter for management HA. Promoting secondaryElastiCenter

Separate ZIL and Cache device Configure separate ZIL and cache device for each Pool.

Block Length for iSCSI LUNs Ensure that the iSCSI LUNs consumed by ESX clients haveblockLength=512

Reducing Dell R720 boot time How to reduce the boot time? Optimizing the boot time forDell PowerEdge R720/Dell

PowerEdge R820

Backing up ElastiCenter andNode configurations

How to backup and restore ElastiCenter and Node con-figurations

Backing up configurations

Storage naming conventions Rules for naming storage components Storage naming conventions

Promoting secondary ElastiCenter and backing up configurations

Setting up and promoting secondary ElastiCenter

The following sections explain how to setup a secondary ElastiCenter (secondary setup) to ensure backup in the case your primaryElastiCenter (primary setup) fails. If a failure occurs, you can promote your secondary setup.

As shown in diagram 1, the primary ElastiCenter manages andmonitors multiple Storage Nodes spread across datacenters. All con-figurations on primary ElastiCenter are replicated on the secondary. In case the primary ElastiCenter fails, the secondary can be pro-moted as shown in diagram 2.

302

http://www.docs.cloudbyte.com/sas-multipathing/


http://www.docs.cloudbyte.com/knowledge-base-articles/optimizing-the-boot-time-for-dell-poweredge-r720dell-poweredge-r820/

http://www.docs.cloudbyte.com/backing-up-elasticenter-configurations/

http://www.docs.cloudbyte.com/storage-naming-convensions/


Diagram 1:

Diagram 2:

Important:The primary and secondary setups cannot be used at the same time. Ensure that only the database service isoperational on the secondary setup. Promote the secondary setup only when the primary setup is down.

Prerequisites

l The primary and secondary setups are on same subnet

l ElastiCenter and Node are installed separately

303


Setting up the secondary ElastiCenter

The procedures involve action at both the primary and secondary setups.

Note: If you copy and paste the commands in the section, ensure that they appear correctly on the CLI.

1. (On both the setups) Log in as root and access (over SSH or directly) the command prompt on themachine where theElastiCenters are installed.

2. (On the secondary setup) Run the following command: service rabbitmq onestop

3. (On the secondary setup) Run the following command to ensure that the RabbitMQ service is stopped: service rab-bitmq onestatus

4. (On secondary setup) Run the following command to ensure that no Java process is running:ps –auxwww | grep java

5. If any of the Java processes or RabbitMQ service is running, kill it using the following command: kill -9 PID_of_pro-cess

6. Else, ensure the following:

l Size of/tpoolfolder is same on both primary ElastiCenter and secondary ElastiCenter using the following command:zpool list

l /folder in secondary ElastiCenter database has enough space for backup. This means the available space in / dir-ectory on the secondary ElastiCenter must be greater than the used space in /tpool directory on the primaryElastiCenter.

7. (Optional) On the secondary setup, start themysql service using the command: service mysql -server onestart .

8. (On the secondary setup) Run the script /usr/local/cb/bin/cb_config_slave.sh .


Select your choice:1) Configure secondary ElastiCenter.2) Promote secondary to primary ElastiCenter.3) Exit1Here, you have to select 1 as highlighted.

You have selected to configure secondary ElastiCenter.Do you want to continue? (Yes/No) YesHere, you have to select Yes as highlighted.

Specify primary ElastiCenter's IP address:20.10.97.1

Wait for the script to execute.Stopping Tomcat and RabbitMQ services...Tomcat and RabbitMQ services stopped.Restarting MySql service...MySql service started.Starting backup of primary ElastiCenter...Primary Elasticenter backup completed.Starting the setting up of secondary ElastiCenter...Setting up the secondary ElastiCenter completed.Syncing reports...

Promoting the secondary setup

Before you proceed

If the primary setup is down, try the following before promoting the secondary setup:

304


l Verify the error logs in java/Rabbitmq/Mysql or the service status. For example, service mysql-serveronestatus.

l Restart the service that is not running.

If downtime prevails despite the service restart, promote the secondary setup.

Procedures

Note: If you copy and paste the commands in the section, ensure that they appear correctly on the CLI.

1. Power off the primary ElastiCenter.

2. (On the secondary ElastiCenter console) Run the script /usr/local/cb/bin/cb_config_slave.sh .


Select your choice:1) Configure secondary ElastiCenter.2) Promote secondary to primary Elasticenter.3) Exit2Here, you have to select 2 as highlighted.You have selected to promote secondary Elasticenter to primary Elasticenter.Do you want to continue ? (Yes/No)YesHere, you have to select Yes as highlighted.

Dropped secondary ElastiCenter.

Starting RabbitMQ and Tomcat services..Promoted secondary Elasticenter to primary Elasticenter. Check catalina logs for errors(if any).

4. (On all the ElastiStor consoles) Modify the /usr/local/agent/listener/cbc.conf file by changing the IP address ofElastiCenter. Specify the IP address of secondary ElastiCenter.

5. (On the ElastiStor console) Run the following command to restart all the python/cbc services on the Node: service -l| grep cbc | grep -v cbc_ec | xargs -I {} service {} onerestart

Also, ensure that the ElastiCenter is accessible to the Nodes with same status as prior to the promotion.

Note: During the promotion process, if a Node is offline or is powered down, thenmanually start the cbc_monitor_hajobs service and point it to the promoted ElastiCenter.

Backing up configurations

How to backup and restore ElastiCenter configuration

By default, CloudByte schedules a backup everyday. You canmodify the settings using the following options in Global Settings:

l devman.backup.files.retain

l devman.backup.path

l devman.backup.scheduling.interval.

But, in the case you need to do amanual backup, follow the steps:

305


Backing up

1. In the ElastiCenter server, at command prompt, run the following: cd /usr/local/cb/bin/

2. Run the following command to create backup file: ./backup.sh. The backup file is created in the /var/devman/backups(unless youmodify the path). The following is a sample file name: backup_Jul26_1374815679.tar.bz2.

3. Copy the file and store it elsewhere.

Restoring

1. Copy the backed up TAR file to a location on ElastiCenter, for example /var.

2. Go to the directory /usr/local/cb/bin and then run the following command: ./restore.sh <path of backedup file>.For example ./restore.sh /var/backup_Jul26_1374815679.tar.bz2.

The configuration is restored.

How to collect logs

Do the following to collect the logs of the datacenter (both Nodes and ElastiCenter):

1. Log in as root.

2. In the ElastiCenter server, at command prompt, run the following command: cd /usr/local/cb/bin/.

3. Run the following command: python logCollector.py

4. At prompt, specify the root password for all Nodes from which you want to gather the logs.

How to backup and restore Node configuration

Backing up

By default, CloudByte schedules a daily backup in the location /cf/conf. CloudByte recommends you tomove this backup dailyto a safe location. Either schedule a cron job or manually copy the files from the conf directory.

You can backup configurations manually by doing the following:

1. cp /usr/local/agent/cb* .

2. cp -r /cf/conf .

Restoring

1. Reinstall the Node with the same configurations (for example, same IP address, Hostname, and ElastiCenter details) as inthe previous installation.

2. Copy the following backed up files to the reinstalled Node:

l config.xml file to the location /cf/conf/config.xml

l haconfig.xml file to the location /cf/conf/haconfig.xml

l ipmi.conf file to the location /usr/local/agent/listener/ipmi.conf

l cbc_node_id file to the location /usr/local/agent/cbc_node_id

l cbd_node_id file to the location /usr/local/agent/cbd_node_id

3. After copying all the files, reboot the Node.

306


IOPS conversion matrix

The following sections provide details of the IOPS conversionmatrix as implemented in CloudByte ElastiStor. The conversionmat-rix helps you understand the penalty enforcement factors associated with the QoS.

Assume that, as a service provider, you have assigned a client a specific number of IOPS. Depending on the work load pattern, tomaintain guaranteedQoS, youmight have to increase or decrease the IOPS. Based on the standard QoS characteristics CloudByteautomatically calculates the equivalent of your load.

For example, in a 100% READ scenario, there is a surplus IOPS owing to the 0%WRITE. Similarly in a 100 percentWRITE scenario, there is a deficit of IOPS. This demands a check and balance in the IOPS provisioning which is taken care of byCloudByte.

Since it is imperative that IOPS provisioning conform to a specific matrix, you are provided with a set of Global Settings optionswhich you can tweak tomaintain and optimize the performance. Note that you can also optimize the performance at per Pool levelalso.

Measuring storage performance

Prior to reading the following sections, CloudByte recommends you to read the following white paper that provides an introduction toMeasuring Storage Performance.

Concepts

IOPS Description

Standard IOPS This is the benchmarked IOPS. The details are as follows: 80 READ/20 WRITE, and 4K block size. This acts asthe frame of reference for QoS in CloudByte ElastiStor.

Operating IOPS IOPS under a given workload that can be measured against the standard IOPS.

Pool IOPS The standard IOPS that can be delivered by a given pool.

Provisioned IOPS . The standard IOPS provisioned to a volume

Illustration

Consider a Pool assigned with 1300 Standard IOPS. There are three Storage Volumes with the following conditions:

Storage Volume Provisioned IOPS Block Size Workload (READ/WRITE) Operational IOPS

Volume 1 500 4K 80% /20% 500

Volume 2 200 4K 100% /0% 170

Volume 3 400 4K 0% /100% 600

In the case of VSM 3whereWRITEs aremore, operational IOPS exceeds the standard IOPS which demands the implementation ofpenalty enforcement.

Penalty enforcement options in Global Settings

The following options in the Global Settings are used to enforce IOPS penalty:

307

http://www.cloudbyte.com/docs/Whitepaper_CloudByte_Measuring-Storage-Performance.pdf


Settings Description

qos.penalty.factor If Yes, penalty is applied based on the criteria explained above.

qos.penalty.enforcement.timeinterval Seconds after which penalty enforcement should refresh.

read.multiplication.factor For details, see the following sections

cache.multiplication.factor For details, see the following sections.

Enabling penalty enforcement factors at the Pool level1. In the Pools page, select the Pool for which you want to enforce IOPS penalty.

2. In the actions pane, click Settings and then click Edit.

3. Select Enable for the option Penalty Enforcement Factor.

4. Click Save.

Deciding the penalty

The following workflow helps you understand the penalty factors:

308


Calculating the operating IOPSStandard IOPS

1000@80%Read, 80%Write, 4K block size

Read/write variation

CloudByte equates oneWRITE to four READS. You can tune this on per pool basis if required.

Calculation

If standard IOPS are 100 at 80 READS/20WRITES, the following is the formula to convert the operating IOPS into standard IOPS:

Oper-atingIOPS

AtREAD/WRI-

TE

Equivalent Standard IOPS Throttle at(IOPS)

100 100/0 80 READS +20 READS (as READs occur in place of WRITEs). Here you convert 20 WRITEs toREADS. That is, 20x4=80 READs. The total standard IOPS that can be consumed is 80 + 80 =160 IOPS.

160

100 80/20 80 READS + 20 READS 100

100 60/40 60 READs + 40 READs (as WRITES occur in place of READs). Here you convert 20 READs toWRITEs. 20 READs are equal to 20/4=5 WRITEs. The total standard IOPS that can be con-sumed is 80+5 = 85 IOPS.

85

309


100 40/20 40 READs + 60WRITEs (as WRITEs occur in place of READs). Here you convert 40 READs toWRITES. 40 READs are equal to 40/4=10 WRITEs. The total standard IOPS that can be con-sumed is 40+10 = 50 IOPS.

50

100 20/80 20 READs + 80 WRITEs (as WRITEs occur in place of READs). Here you convert 60 READs toWRITEs. 60 READs are equal to 60/4=15 WRITEs. The total standard IOPS that can be con-sumed is 20+15 = 35 IOPS.

35

100 0/100 0 READs + 100 WRITEs (as WRITEs occur in place of READs). Here you convert 80 READs toWRITEs. 80 READs are equal to 80/4=20 WRITEs. The total standard IOPS that can be con-sumed is 0+20 = 20 IOPS.

20

READ/WRITE variation and block size variation

The calculation considers the following assumptions:

l X operating IOPS are converted to X1 after considering the block size variation

l X1 IOPS are converted to X2 after considering READ/WRITE variation

l Throttling happens at X2 IOPS

310

CHAPTER XIII Glossary


Terms Description

Account An entity with specific storage needs. Account represents a department of an Enterprise or a customer of aCloud Service Provider.

Alerts Notifications that demand action. Alerts can be on a range of issues such as network cable pullout or degradationof a Pool.

Asynchronous rep-lication

The process of writing data to local storage and then queuing it up for replication on remote storage at assignedintervals. If the local storage fails before the data is written in remote storage, then replication fails.

Auto-Provisioning ElastiStor can automate Node selection and resource allocation if you provide the required SLA/QoS parameters.Sensing the level of storage performance and capacity the application requires, ElastiStor automatically allots theright Pool.

Block Size The size of a block in data storage and file systems. Block is a sequence of bytes or bits, having a nominal size.

CIFS For details, see http://technet.microsoft.com/en-us/library/cc939973.aspx

HA Group Houses a group of Nodes. HA Group enables takeover during failures. Workload of a failed Node is assigned toanother Node in the same HA Group based on availability and performance requirements.

Dashboard A graphical representation of the key performance indicators. It presents an overview of all the Sites inElastiCenter in an organized manner. You can update information instantaneously.

Delegated Admin-istration

The process of distributing administrative tasks and content in the portal to dedicated administrators. Usingaccess control lists, you determine the actions administrators take on various resources. Delegated admin-istration roles are mapped to administrative functions using security policies. You can create and maintain sep-arate administration capabilities depending on your needs.

Disaster Recovery Disaster implies unplanned outage. Your storage systemmust be capable of effective backup and recovery sothat there is no damage to the data that is handled in the case of any adversity.

Disk Array Popularly called JBODs, disk array is a set of disks clubbed to single disk shelf.

Disk Type The type of disk supported by CloudByte ElastiStor. The supported disk types are SAS, SATA, and SSD.

Domain Search Searching the DNS to which you want to add a controller.

Event Any storage activity. ElastiStor notifies events based on the configuration.

FC (Fibre Channel) A high-speed interconnect between controllers and back-end disk enclosures. FC components include HBAs,hubs, and switches. The term also refers to a high-speed, fully duplex serial communication protocol permittingdata transfer up to 10GB per second.

FTP (File TransferProtocol)

An application layer protocol used to transfer bulk-volumes of files between machines or hosts.

Global Settings Lets the administrator configure options that control ElastiStor.

Grace Provisioning of Pool's unused IOPS/Throughput to Storage Volumes based on their performance requirements.

High-Availability High Availability minimizes/prevents the consequences of downtime in the storage setup. A typical High Avail-ability environment comprises two setups: one active while the partner remains dormant. In a downtime, the pass-ive setup takes over the storage tasks that the active setup was performing. This effects a reversal of states eachsetup was in. When the downtime is over, all events are given back to the original active setup.

Host Physical machines on which you configure ElastiStor.

311


Terms Description

Initiator Group When an initiator group is associated with a LUN, only initiators from that group can access the LUN. Initiators areidentified by unique addressing methods.

IOPS (Input/Out-put operations persecond)

The number of READs or WRITEs per second. It is used to measure the performance of storage solution. TotalIOPS is the average input/output operations per second, Read IOPS is the average READ operations persecond, andWrite IOPS is the average WRITE operations per second.

iSCSI Internet Small Computer Systems Interface is a client-server protocol used to enable the transfer of SCSI packetsover a TCP/IP (Ethernet) network.

Lagg Interface Allows integration of multiple network interfaces with a single virtual interface. This helps in fault-tolerance andhigh-speed multi-link throughput.

LUN (Logical UnitNumber)

Unique address to identify a logical unit. Logical unit is the conceptual subunit of a storage disk or set of disks.

Mirrored Disks A High Availability setup where you have a different set of disks on each node. Active disks are rep-licated/mirrored on the partnering node.

Mirroring Storing duplicate copies of data on one or more sets of additional Storage Volumes, apart from the set of Stor-age Volume that stores the original data.

Monitoring Provides an intuitive visualization of storage statistics. Monitoring foregrounds relevant back end activities in aneasy user interface for the administrator.

Network Failure Indicates that the system is unable to reach the network. For example, due to physical port failure.

NFS A distributed file system protocol that lets clients access files over a network in the same way they access localstorage.

Physical Infra-structure

Real entities pertaining to the hardware.

Pool Physical storage devices on which you store data.

Provisioning A set of steps to allocate storage to Hosts and applications. For example, creating Storage Volumes in Pools.

QoS (Quality ofService)

Lets you easily commission storage with guaranteed IOPS at various levels. You can decide to configure QoS at aPool level, VSM level, or Storage Volume level.

Report Detailed information related to configuration. Use Configuration Reports for component configuration details.Dynamic Reports provide QoS statistics.

Rest APIs Every action performed at ElastiStor admin console translates into a REST based API call in the back end. Thisenables administrators to fully manage ElastiStor from their portals. CloudByte plugin for VMware vCenter (alsobased on REST API) enables storage management from vCenter.

Shared Disks A High Availability setup where you have a common set of disks shared between primary and secondary nodes.The disks can be either hard disks or iSCSI disks.

Shared Storage Storage devices that are connected to and used by two or more systems.

Site Represents a data center or part of a data center. Building storage infrastructure starts with the creation of Site.It is the fundamental infrastructure unit where your controller, storage, and network infrastructure reside.

Snapshot Copy of a Storage Volume created at specific intervals.

Storage Capacity The overall disk space (and not the free disk space).

Striping Writes data across two or more disks in an array but utilizes only a part of the space on each disk. The amount ofspace used by stripe on each disk is always the same. A virtual disk may write several stripes on the same set ofdisks in the array.

312


Terms Description

Synchronous Rep-lication

The process of writing the same data to two separate volumes, stored on different physical hardware, before theapplication receives an I/O complete message. This results in both the volumes being always in sync.

Takeover The process of taking over the storage control activity from a node. Typically, when there is a downtime on theprimary node, the secondary node takes over the control of storage.

Thin Provisioning The process of optimizing efficiency of using the available storage space. It is the allocation of disk storage spaceflexibly among multiple users, based on the minimum space each user requires at any given time.

Throughput Amount of work a system can perform at specific intervals. It is also the amount of data moved successfully fromone place to another in a given time period.

Virtual StorageMachine(VSM )

Virtual storage controller capable of hosting Storage Volumes and LUNs.

VSM Architecture Unique to CloudByte ElastiStor, this is a patented architecture. In a Storage Controller, each application is fullyisolated across all storage stack levels and unified under a Virtual Storage Machine (VSM). VSM isolation allowscontroller resources allotted to applications to be easily monitored, controlled, and provisioned.

VMware vSphereclient

A Microsoft Windows desktop application that enables you to access VMware ESXi and VMware vCenter.

ZFS (Zettabyte FileSystem)

ZFS is a 128-bit file system that provides features such as data integrity verification, disk management, snapshots,and transactional operations.

313

CloudByte ElastiStor Guide

Documents