Dell EMC Host Connectivity Guide for Windows

Dell EMC Host
Connectivity Guide
for Windows
P/N 300-000-603
REV 59
This document is not intended for audiences in China, Hong Kong, and Taiwan.
Copyright © 2015 - 2017 Dell Inc. or its subsidiaries. All rights reserved.
Published May 2017
Dell believes the information in this publication is accurate as of its publication date. The information is subject to change without
notice.
THE INFORMATION IN THIS PUBLICATION IS PROVIDED "AS IS." EMC CORPORATION MAKES NO REPRESENTATIONS OR
WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS
IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.
Dell, EMC2, EMC, and the EMC logo are registered trademarks or trademarks of Dell Inc. or its subsidiaries. All other trademarks
used herein are the property of their respective owners..
For the most up-to-date regulator document for your product line, go to Dell EMC Online Support (https://support.emc.com).
2
Dell EMC Host Connectivity Guide for Windows
CONTENTS
Preface........................................................................................................................................ 9
Chapter 1
General Procedures and Information
General Windows information.................................................................... 14
Terminology........................................................................................ 14
Utilities and functions ......................................................................... 14
Windows environment ............................................................................... 15
Hardware connectivity........................................................................ 15
Booting Windows from external storage.................................................... 16
Boot-from-SAN .................................................................................. 16
Benefits of boot-from-SAN ................................................................ 16
Boot-from-SAN configuration restrictions.......................................... 16
Risks of booting from the storage array.............................................. 17
How to determine I/O latency and load on the boot LUN ................... 17
Configuring Unity and VNX series systems for boot from SAN ........... 18
SAN Booting a Windows Host to a Unity array .......................................... 19
Prerequisites ...................................................................................... 19
Configure host connections................................................................ 19
Create a LUN and configure to the host .................................................... 23
Microsoft Windows Failover Clustering ..................................................... 28
Chapter 2
iSCSI Attach Environments
Introduction.............................................................................................. 30
Terminology....................................................................................... 30
Software............................................................................................ 30
Boot device support .......................................................................... 30
Windows 2008 R2 iSCSI Initiator manual procedure.................................. 31
Windows 2008 R2 iSCSI Initiator cleanup.......................................... 35
Using MS iSNS server software with iSCSI configurations........................ 38
iSCSI Boot with the Intel PRO/1000 family of adapters............................. 39
Preparing your storage array for boot................................................. 39
Post installation information ............................................................... 41
Notes on Microsoft iSCSI Initiator............................................................. 44
iSCSI failover behavior with the Microsoft iSCSI initiator................... 44
Microsoft Cluster Server ................................................................... 60
Boot................................................................................................... 60
NIC teaming........................................................................................ 61
Using the Initiator with PowerPath..................................................... 61
Commonly seen issues....................................................................... 65
Chapter 3
Virtual Provisioning
Virtual Provisioning on Symmetrix .............................................................
Terminology........................................................................................
Management tools ..............................................................................
Thin device .........................................................................................
Implementation considerations..................................................................
Over-subscribed thin pools .................................................................
70
71
72
72
74
74
Dell EMC Host Connectivity Guide for Windows
3
Contents
Thin-hostile environments .................................................................. 75
Pre-provisioning with thin devices in a thin hostile environment......... 75
Host boot/root/swap/dump devices positioned on
Symmetrix VP (tdev) devices ............................................................. 76
Cluster configurations ........................................................................ 76
Operating system characteristics .............................................................. 78
Chapter 4
Windows Host Connectivity with Dell EMC VPLEX
Dell EMC VPLEX ...................................................................................... 80
Prerequisites ............................................................................................. 81
Host connectivity ...................................................................................... 82
Configuring Fibre Channel HBAs ............................................................... 83
Setting queue depth and execution throttle for QLogic ...................... 83
Setting queue depth and queue target for Emulex.............................. 88
Windows Failover Clustering with VPLEX.................................................. 91
Setting up quorum on a Windows 2012/2012 R2 Failover Cluster
for VPLEX Metro or Geo clusters .............................................................. 92
Configuring quorum on Windows 2008/2008 R2 Failover Cluster
for VPLEX Metro or Geo clusters ............................................................. 96
VPLEX Metro or Geo cluster configuration........................................ 96
Prerequisites ...................................................................................... 97
Setting up quorum on a Windows 2008/2008R2 Failover Cluster
for VPLEX Metro or Geo clusters ....................................................... 97
Chapter 5
Dell EMC PowerPath for Windows
PowerPath and PowerPath iSCSI............................................................ 104
PowerPath for Windows.......................................................................... 105
PowerPath and MSCS...................................................................... 105
Integrating PowerPath into an existing MSCS cluster ...................... 105
PowerPath verification and problem determination ................................. 108
Problem determination ...................................................................... 110
Making changes to your environment ................................................ 113
PowerPath messages ........................................................................ 113
Chapter 6
Microsoft Native MPIO and Hyper-V
Native MPIO with Windows Server 2008/Windows Server 2008 R2........ 116
Support for Native MPIO in Windows Server 2008 and
Windows Server 2008 R2 .................................................................. 116
Configuring Native MPIO for Windows 2008 Server Core and
Windows 2008 R2 Server Core.......................................................... 116
Native MPIO with Windows Server 2012 ................................................. 120
Support for Native MPIO in Windows Server 2012 ........................... 120
Configuring Native MPIO for Windows Server 2012 ......................... 120
Known issues .......................................................................................... 125
Hyper-V................................................................................................... 126
Appendix A
Persistent Binding
Understanding persistent binding ........................................................... 128
Methods of persistent binding .......................................................... 130
4
Dell EMC Host Connectivity Guide for Windows
Contents
Appendix B
Dell EMC Solutions Enabler
Dell EMC Solutions Enabler .................................................................... 132
References ....................................................................................... 132
Appendix C
Veritas Volume Management Software ......................................... 133
Dell EMC Host Connectivity Guide for Windows
5
Contents
6
Dell EMC Host Connectivity Guide for Windows
FIGURES
FIGURES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Four paths ................................................................................................................ 42
PowerPathAdmin ..................................................................................................... 43
Advanced Settings dialog box .................................................................................. 44
Single iSCSI subnet configuration ........................................................................... 45
Multiple iSCSI subnet configuration ......................................................................... 51
iSCSI Initiator Properties dialog box ......................................................................... 62
Log On to Target dialog box ..................................................................................... 63
Advanced Settings dialog box .................................................................................. 63
Four paths ................................................................................................................ 64
Virtual Provisioning on Symmetrix ............................................................................ 70
Thin device and thin storage pool containing data devices ....................................... 73
VPLEX Metro cluster configuration example ........................................................... 96
PowerPath Administration icon .............................................................................. 108
PowerPath Monitor Taskbar icons and status ........................................................ 108
One path ................................................................................................................ 109
Multiple paths ......................................................................................................... 110
Error with an Array port .......................................................................................... 112
Failed HBA path ...................................................................................................... 113
MPIO Properties dialog box .................................................................................... 122
Original configuration before the reboot ................................................................ 129
Host after the reboot ............................................................................................. 129
Dell EMC Host Connectivity Guide for Windows
7
8
Dell EMC Host Connectivity Guide for Windows
PREFACE
As part of an effort to improve and enhance the performance and capabilities of its
product line, Dell EMC from time to time releases revisions of its hardware and
software. Therefore, some functions described in this document may not be supported
by all revisions of the software or hardware currently in use. For the most up-to-date
information on product features, refer to your product release notes.
If a product does not function properly or does not function as described in this
document, contact your Dell EMC Representative.
This guide describes the features and setup procedures for Windows 2016, 2012 R2,
Windows 2012, and Windows 2008 R2 host interfaces to Dell EMC storage arrays over
Fibre Channel or iSCSI.
Note: This document was accurate at publication time. New versions of this document
might be released on Dell EMC Online Support. Check to ensure that you are using the
latest version of this document.
Audience
This guide is intended for use by storage administrators, system programmers, or
operators who are involved in acquiring, managing, or operating Dell EMC VMAX™ All
Flash Family, Dell EMC VMAX3™ Family, Dell EMC VMAX Family, Dell EMC Unity™
Family, EMC Unified VNX™ series, Dell EMC XtremIO™, Dell EMC VPLEX™, and host
devices, and Windows Server 2016, Windows Server 2012 R2, Windows Server 2012,
Windows Server 2008, and Windows Server 2008 R2.
Readers of this guide are expected to be familiar with the above storage systems and
the operation of them.
Dell EMC Support
Matrix
For the most up-to-date information, always consult the Dell EMC Simple Support
Matrix (ESM) on E-Lab Interoperability Navigator (ELN).
VMAX All Flash,
VMAX3, VMAX,
Symmetrix™, and
VNX references
Unless otherwise noted:
◆
Any general references to VMAX3 include the VMAX All Flash Family, VMAX3
Family, VMAX Family, and DMX.
◆
Any general references to Unity include any array models in the Unity and Unified
VNX Families.
Related
documentation
For Dell EMC documentation, refer to Dell EMC Online Support.
Conventions used in
this guide
Dell EMC uses the following conventions for notes and cautions.
IMPORTANT
An important notice contains information essential to software or hardware operation.
Note: A note presents information that is important, but not hazard-related.
Dell EMC Host Connectivity Guide for Windows
9
Typographical conventions
Dell EMC uses the following type style conventions in this document.
Normal
Used in running (nonprocedural) text for:
• Names of interface elements, such as names of windows, dialog
boxes, buttons, fields, and menus
• Names of resources, attributes, pools, Boolean expressions, buttons,
DQL statements, keywords, clauses, environment variables,
functions, and utilities
• URLs, pathnames, filenames, directory names, computer names,
links, groups, service keys, file systems, and notifications
Bold
Used in running (nonprocedural) text for names of commands,
daemons, options, programs, processes, services, applications, utilities,
kernels, notifications, system calls, and man pages
Used in procedures for:
• Names of interface elements, such as names of windows, dialog
boxes, buttons, fields, and menus
• What the user specifically selects, clicks, presses, or types
Where to get help
Italic
Used in all text (including procedures) for:
• Full titles of publications referenced in text
• Emphasis, for example, a new term
• Variables
Courier
Used for:
• System output, such as an error message or script
• URLs, complete paths, filenames, prompts, and syntax when shown
outside of running text
Courier bold
Used for specific user input, such as commands
Courier italic
Used in procedures for:
• Variables on the command line
• User input variables
<>
Angle brackets enclose parameter or variable values supplied by the
user
[]
Square brackets enclose optional values
|
Vertical bar indicates alternate selections — the bar means “or”
{}
Braces enclose content that the user must specify, such as x or y or z
...
Ellipses indicate nonessential information omitted from the example
Dell EMC support, product, and licensing information can be obtained as follows.
Dell EMC support, product, and licensing information can be obtained on the Dell EMC
Online Support site as described next.
Note: To open a service request through the Dell EMC Online Support site, you must
have a valid support agreement. Contact your Dell EMC sales representative for details
about obtaining a valid support agreement or to answer any questions about your
account.
10
Dell EMC Host Connectivity Guide for Windows
Product information
For documentation, release notes, software updates, or for information about Dell
EMC products, licensing, and service, go to Dell EMC Online Support (registration
required).
Technical support
Dell EMC offers a variety of support options.
Support by Product — Dell EMC offers consolidated, product-specific information
on the Web at the Dell EMC Support By Product page.
The Support by Product web pages offer quick links to Documentation, White Papers,
Advisories (such as frequently used Knowledgebase articles), and Downloads, as well
as more dynamic content, such as presentations, discussion, relevant Customer
Support Forum entries, and a link to EMC Live Chat.
Dell EMC Live Chat — Open a Chat or instant message session with an Dell EMC
Support Engineer.
eLicensing support
To activate your entitlements and obtain your Symmetrix license files, visit the Service
Center on Dell EMC Online Support, as directed on your License Authorization Code
(LAC) letter that was e-mailed to you.
For help with missing or incorrect entitlements after activation (that is, expected
functionality remains unavailable because it is not licensed), contact your Dell EMC
Account Representative or Authorized Reseller.
For help with any errors applying license files through Solutions Enabler, contact the
Dell EMC Customer Support Center.
If you are missing a LAC letter, or require further instructions on activating your
licenses through the Online Support site, contact Dell EMC's worldwide Licensing
team at licensing@emc.com or call:
◆
North America, Latin America, APJK, Australia, New Zealand: SVC4EMC
(800-782-4362) and follow the voice prompts.
◆
EMEA: +353 (0) 21 4879862 and follow the voice prompts.
We'd like to hear from you!
Your suggestions will help us continue to improve the accuracy, organization, and
overall quality of the user publications. Send your opinions of this document to:
techpubcomments@emc.com
Dell EMC Host Connectivity Guide for Windows
11
12
Dell EMC Host Connectivity Guide for Windows
CHAPTER 1
General
Procedures
and
Information
This chapter provides general procedures and information about Windows hosts.
◆
◆
◆
◆
General Windows information ..................................................... 14
Windows environment................................................................. 15
Booting Windows from external storage...................................... 16
Microsoft Windows Failover Clustering ...................................... 28
General Procedures and Information
13
General Procedures and Information
General Windows information
This section provides information that is common to all supported versions of
Windows. Read the entire section before proceeding to the rest of the chapter.
Terminology
You should understand these terms:
◆
Free space—An unused and unformatted portion of a hard disk that can be
partitioned or subpartitioned.
◆
Partition—A portion of a physical hard disk that functions as though it were a
physically separate unit.
◆
Volume—A partition or collection of partitions that have been formatted for use
by a file system. A volume is assigned a drive letter.
◆
Primary partition—A portion of a physical disk that can be marked for use by an
operating system. A physical disk can have up to four primary partitions. A primary
partition cannot be subpartitioned.
Utilities and functions
Here are some Windows functions and utilities you can use to define and manage
VMAX3, Unity series, Dell EMC Unified systems, and XtremIO systems. The use of
these functions and utilities is optional; they are listed for reference only:
14
◆
Disk Manager—Graphical tool for managing disks; for example, partitioning,
creating, and deleting volumes.
◆
Registry Editor—Graphical tool for displaying detailed hardware and software
configuration information. Not normally part of the Administrative Tools group, the
registry editor, REGEDT.EXE, is in the Windows \system32 subdirectory.
◆
Event Viewer—Graphical tool for viewing system or application errors.
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
Windows environment
This section lists Fibre Channel support information specific to the Windows
environment.
For more information, refer to the appropriate chapter:
◆
Chapter 2, ”iSCSI Attach Environments”
Hardware connectivity
Refer to the Dell EMC Simple Support Matrix or contact your Dell EMC representative
for the latest information on qualified hosts, host bus adapters, and connectivity
equipment.
Dell EMC does not recommend mixing HBAs from different vendors in the same host.
Windows environment
15
General Procedures and Information
Booting Windows from external storage
Windows hosts have been qualified for booting from Dell EMC array devices interfaced
through Fibre Channel as described under Boot Device Support in the Dell EMC Simple
Support Matrix. Refer to the appropriate Windows HBA guide, available on Dell EMC
Online Support, for information on configuring your HBA and installing the Windows
operating system to an external storage array:
◆
EMC Host Connectivity with Emulex Fibre Channel Host Bus Adapters (HBAs) and
Converged Network Adapters (CNAs) in the Windows Environment 7
◆
EMC Host Connectivity with QLogic Fibre Channel and iSCSI Host Bus Adapters
(HBAs) and Converged Network Adapters (CNAs) in the Windows Environment
◆
EMC Host Connectivity with Brocade Fibre Channel and Host Bus Adapters
(HBAs) and Converged Network Adapters (CNAs) in the Windows Environment
Boot-from-SAN
Although Windows servers typically boot the operating system from a local, internal
disk, many customers want to utilize the features of VMAX3, Unity, Unified VNX, and
XtremIO to store and protect their boot disks and data. Boot-from-SAN enables
VMAX3, Unity, Unified VNX, and XtremIO to be used as the boot disk for your server
instead of a directly-attached (or internal) hard disk. Using a properly configured Fibre
Channel HBA, FCoE CNA, or blade server mezzanine adapter connected and zoned to
the same switch or fabric as the storage array, a server can be configured to use a LUN
presented from the array as its boot disk.
Benefits of boot-from-SAN
Boot-from-SAN can simplify management in the data center. Separating the boot
image from each server allows administrators to leverage their investments in Dell
EMC storage arrays to achieve high availability, better data integrity, and more efficient
storage management. Other benefits can include:
◆
Improved disaster tolerance
◆
Reduced total cost through diskless servers
◆
High-availability storage
◆
Rapid server repurposing
◆
Consolidation of image management
Boot-from-SAN configuration restrictions
Refer to the Dell EMC Simple Support Matrix for any specific boot-from-SAN
restrictions since this guide no longer contains restriction information. The information
in the Dell EMC Simple Support Matrix supersedes any restriction references found in
previous HBA installation guides.
16
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
Risks of booting from the storage array
When using the storage array as a boot disk, Dell EMC recommends that you shut
down the host server during any maintenance procedures that could make the boot
disk unavailable to the host.
IMPORTANT
Microsoft Windows operating systems use virtual memory paging files that reside on
the boot disk. If the paging file becomes unavailable to the memory management
system when it is needed, the operating system will crash with a blue screen.
Any of these events could crash a system booting from the storage array:
◆
Lost connection to array (pulled or damaged cable connection)
◆
Array service/upgrade procedures, such as on-line microcode upgrades and/or
configuration changes
◆
Array failures, including failed lasers on Fibre Channel ports
◆
Array power failure
◆
Storage Area Network failures, such as Fibre Channel switches, switch
components, or switch power failures
◆
Storage Area Network service/upgrade procedures, such as firmware upgrades or
hardware replacements
Note: Dell EMC recommends moving the Windows virtual memory paging file to a
local disk when booting from the storage array. Consult your Windows manual for
instructions on how to move the paging file.
How to determine I/O latency and load on the boot LUN
The current restrictions for boot-from-array configurations listed in the Dell EMC
Simple Support Matrix represent the maximum configuration that is allowed using
typical configurations. There are cases where your applications, host, array, or SAN
may already be utilized to a point when these maximum values might not be achieved.
Under these conditions, you may wish to reduce the configuration from the maximums
listed in the Dell EMC Simple Support Matrix for improved performance and
functionality.
Here are some general measurements than can be used to determine if your
environment might not support the maximum allowed boot-from-array configurations:
◆
Using the Windows Performance Monitor, capture and analyze the Physical Disk
and Paging File counters for your boot LUN. If response time (sec/operation), or
disk queue depth seem to be increasing over time, you should review any additional
loading that may be affecting the boot LUN performance (HBA/SAN saturation,
failovers, ISL usage, and so forth).
◆
Use available Array Performance Management tools to determine that the array
configuration, LUN configuration and access is configured optimally for each host.
Booting Windows from external storage
17
General Procedures and Information
Possible ways to reduce the load on the boot LUN include:
◆
Move application data away from the boot LUN.
◆
Reduce the number of LUNs bound to the same physical disks.
◆
Select an improved performance RAID type.
◆
Contact your Dell EMC support representative for additional information.
Configuring Unity and VNX series systems for boot from SAN
By default, Unity series and Unified VNX storage systems are configured with all of the
proper settings that a Windows server requires for a successful boot from SAN. Unity
and Unified VNX series storage systems have two storage processors (SPs) which
allow for highly available data access even if a single hardware fault has occurred. In
order for a host to be properly configured for high availability with boot-from-SAN, the
HBA BIOS should have connections to both SPs on the Unity and Unified VNX system.
At the start of the Windows boot procedure, there is no failover software running. HBA
BIOS, with a primary path and secondary path(s) properly configured (with access to
both SPs), will provide high availability while booting from SAN with a single hardware
fault.
IMPORTANT
Dell EMC strongly recommends using failover mode 4 (ALUA active/active) when
supported, as ALUA will allow I/O access to the boot LUN from either SP, regardless of
which SP currently owns the boot LUN.
Failover mode 1 is an active/passive failover mode. I/O can only successfully complete
if it is directed to the SP that currently owns the boot LUN. If HBA BIOS attempts to
boot from a passive path, BIOS will have to time out before attempting a secondary
path to the active (owning) SP, which can cause delays at boot time. Using ALUA
failover mode whenever possible will avoid these delays.
To configure a host to boot from SAN, the server needs to have a boot LUN presented
to it from the array, which requires that the WWN of the HBA(s) or CNA(s), or the iqn
of an iSCSI host, be registered.
In configurations where a server is already running Windows and is being attached to a
Unity series and Unified VNX systems, the Dell EMC Unisphere™/Navisphere™ Agent
would be installed on the server. This agent would automatically register the server’s
HBA(s) WWNs on the array. In boot-from-SAN configurations where the OS is going to
be installed on the Unity series and Unified VNX series, there is no agent available to
perform the registration. Manual registration of the HBA WWNs is required in order to
present a LUN to the server for boot. The following SAN Booting a Windows Host to a
Unity array includes this procedure.
For instructions on how to register a host to boot from an iSCSI-based SAN, refer to
Chapter 2, ”iSCSI Attach Environments.”
18
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
SAN Booting a Windows Host to a Unity array
Use the following procedure to add the host to the Unity array, and connect the SAN
Boot LUN to the host in preparation for installing Windows Server.
Prerequisites
Zone the host with one HBA port to one target port of the Unity array. For the OS, it is
necessary to have only one path to the boot LUN. After the operating system is
installed, you can zone and configure multiple paths with the Unity array.
Configure host connections
1. Open the Unisphere Manager connection to the Unity array.
2. Under Access, select Hosts > + Host, as follows:
SAN Booting a Windows Host to a Unity array
19
General Procedures and Information
3. Type a Name and Description and click Next, as follows:
4. Select the Operating System, type the Network Address, and click Next, as
follows:
20
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
5. On the Select iSCSI Initiators page, click Next, as follows:
6. Under Select Fibre Channel Initiators > Auto-Discovered Initiators, select the
Initiator WWN that the host will use to access storage resources.
Note: If you do not find the initiator you want in the list, click Create Initiator to
manually add an initiator, and then select it from the list of Manually Created
Initiators.
Click Next when you are finished, as follows:
SAN Booting a Windows Host to a Unity array
21
General Procedures and Information
7. Review the host configuration and click Finish, as follows.
22
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
Create a LUN and configure to the host
This example uses Block storage.
1. Open Unisphere Manager
2. Under Storage, select Block > Launch the Create LUN Wizard by selecting the +
sign under the LUNs.
3. Type a Name and Description and click Next, as follows:
4. In the Create a LUN dialog box, select a storage Pool, Tiering Policy, LUN Size,
and Host I/O Limit, and then select Next, as follows:
Create a LUN and configure to the host
23
General Procedures and Information
5. In the Select Host Access dialog box, select the host name that was previously
configured, and click OK, as follows:
6. Under Configure Access, select the hosts that can access the storage resource.
For block-based storage, you can configure each host to access the storage
resource, snapshots of the storage resource, or both. Click Next when you are
finished, as follows:
7. Under Configure Snapshot Schedule, select Enable Automatic Snapshot
Creation, and select a Snapshot schedule from the list.
24
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
The following example shows default protection of every day at 3:00 AM for 2
days. You can also select New Schedule to create a schedule that is not in the list.
Click Next when you are finished:
8. To set the replication mode and recovery point objective (RPO), select Enable
Replication and follow the directions at the right side of the following dialog box:
Notes:
• To create a remote replication interfaces, navigate to the Data Protection >
Replication > Interfaces.
• o set up replication using RecoverPoint, go back to step 6, Configure Access,
and configure the RecoverPoint host to access the storage resource.
Create a LUN and configure to the host
25
General Procedures and Information
9. Review your settings in the following Summary. Click Back to change any
settings, and click Finish to accept them:
The following Results screen confirms the completion of the preparation steps for
installing Windows. After the installation, you can zone and configure secondary
paths and a second HBA port:
IMPORTANT
Make note of the host name you chose during the manual registration process. If you
install the Unisphere/Navisphere host agent on your Windows server after installation,
you must ensure that your Windows server is given the same name that you used
during registration. If the name is different, and you install the Unisphere/Navisphere
host agent, your registration on the VNX series or Unity system could be lost and your
server could lose access to the boot LUN and crash.
Manually registering an HBA WWN/WWPN
Your server HBA(s) WWN/WWPN can also be registered manually if it has already
been zoned and logged into the array port. To manually register an HBA WWN/WWPN
that is already logged into the array, refer to the vendor HBA documentation.
26
Dell EMC Host Connectivity Guide for Windows
General Procedures and Information
Using Naviseccli to create an initiator record or manually register an
HBA WWN/WWPN
The secure Navisphere command line utility naviseccli may also be used to create an
initiator record or manually register an HBA WWN/WWPN. All the selections required
in the “Manually registering an HBA WWN/WWPN” examples can be included in a
single naviseccli storagegroup command. Refer to the naviseccli documentation on
Dell EMC Online Support for full details of the switches of the storagegroup
command.
Configuring VMAX3 arrays for boot from SAN
Unlike Unity series and Unified VNX systems, VMAX3 arrays can not be configured
with all of the proper settings a Windows server requires for successful boot from
SAN. Specific VMAX3 director flags (sometimes referred to as director bits) are
required. These flags must be enabled on every port that a Windows server is attached
to. VMAX3 arrays are highly available, with multiple connections (FA ports) for failover
if hardware faults occur. For a host to be properly configured for high availability with
boot from SAN, the HBA BIOS should have connections to at least two connections on
the VMAX3 array.
At the start of the Windows boot procedure, there is no failover software running. HBA
BIOS, with a primary path and secondary path(s) properly configured (to separate FA
ports), will provide high availability while booting from SAN with a single hardware
fault.
To configure a host to boot from SAN, the server needs to have a boot LUN presented
to it from the array. Unlike Unity series and Unified VNX systems, VMAX3 arrays do not
require that an HBA's WWPN be registered. However, VMAX3 storage arrays do
provide LUN masking features that require the HBA WWPN to be validated in the
array’s device-masking database.
Various families of the Dell EMC HYPERMAX OS microcode use different techniques
to enable and configure their LUN-masking features. In order to configure and apply
LUN-masking for your array model, Dell EMC Solutions Enabler software can be used
to issue commands to the VMAX3 array using the Solutions Enabler command line
interface (CLI) to perform LUN-masking on a VMAX3 array.
Refer to the Solutions Enabler Array Controls and Management 8.3.0 CLI User Guide,
located on Dell EMC Online Support, for instruction on using Solutions Enabler CLI to
perform LUN-masking for your Symmetrix model.
Note: It is assumed that your host HBA WWN/WWPN has not yet been zoned to the
Symmetrix array.
Create a LUN and configure to the host
27
General Procedures and Information
Microsoft Windows Failover Clustering
Failover clustering—a Windows Server feature that enables you to group multiple
servers together into a fault-tolerant cluster—provides new and improved features for
software-defined datacenter customers and for many other workloads that run
clusters on physical hardware or in virtual machines.
A failover cluster is a group of independent computers that work together to increase
the availability and scalability of clustered roles (formerly called clustered
applications and services). The clustered servers (called nodes) are connected by
physical cables and by software. If one or more of the cluster nodes fail, other nodes
begin to provide service (a process known as failover). In addition, the clustered roles
are proactively monitored to verify that they are working properly. If they are not
working, they are restarted or moved to another node.
Failover clusters also provide Cluster Shared Volume (CSV) functionality that provides
a consistent, distributed namespace that clustered roles can use to access shared
storage from all nodes. With the Failover Clustering feature, users experience a
minimum of disruptions in service.
Failover Clustering has many practical applications, including:
◆
Highly available or continuously available file share storage for applications such
as Microsoft SQL Server and Hyper-V virtual machines
◆
Highly available clustered roles that run on physical servers or on virtual machines
that are installed on servers running Hyper-V
For more information, refer to Failover Clustering in Windows Server 2016 in the
Microsoft Windows IT Center.
28
Dell EMC Host Connectivity Guide for Windows
CHAPTER 2
iSCSI Attach
Environments
This chapter provides information on the Microsoft iSCSI Initiator and the Microsoft
Cluster Server.
◆
◆
◆
◆
◆
Introduction ............................................................................... 30
Windows 2008 R2 iSCSI Initiator manual procedure ................... 31
Using MS iSNS server software with iSCSI configurations ........ 38
iSCSI Boot with the Intel PRO/1000 family of adapters ............. 39
Notes on Microsoft iSCSI Initiator.............................................. 44
iSCSI Attach Environments
29
iSCSI Attach Environments
Introduction
Microsoft Internet iSCSI Initiator enables you to connect a host computer that is
running Windows Server 2008, Windows Server 2008 R2 Windows Server 2012, and
Windows Server 2012 R2 to an external iSCSI-based storage array through an Ethernet
network adapter. You can use Microsoft iSCSI Initiator in your existing network
infrastructure to enable block-based storage area networks (SANs). SANs provide
iSCSI target functionality without investing in additional hardware.
Terminology
You should understand these terms:
◆
Challenge Handshake Access Protocol (CHAP) — An authentication method
that is used during the iSCSI login in both the target discovery and the normal
login.
◆
iSCSI Network Portal — The host NIC IP address that is used for the iSCSI driver
to create a session with the storage.
Software
Node-names
The Microsoft iSCSI initiator supports an iSCSI target that is configured with a
node-name as the following rules:
◆
Node-names are encoded in UTF8 Character set.
◆
The length of a node-name should be 223 characters or less.
◆
A name can include any of the following valid characters:
• a through z (upper or lower case; uppercase characters are always mapped to
lowercase)
• 0 through 9
• . (period)
• - (dash)
• : (colon)
Refer to the Microsoft iSCSI Initiator x.0 User’s Guide for the complete set of rules for
setting up the valid initiator and target node-names.
Boot device support
The Microsoft iSCSI initiator does not support booting the iSCSI host from iSCSI
storage. Refer to the Dell EMC Simple Support Matrix for the latest information about
boot device support.
30
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Windows 2008 R2 iSCSI Initiator manual procedure
Note: Microsoft iSCSI Initiator is installed natively on Windows Server 2012 R2,
Windows Server 2012, Windows Server 2008 R2, and Windows Server 2008. On these
operating systems, no installation steps are required. For more details and Windows
2012/2012 R2 information, refer to the Microsoft iSCSI Initiator Step-by-Step Guide
document located at technet.microsoft.com.
Prior to configuring the iSCSI initiator, ensure you have decided exactly which NIC will
connect to which target.
For example:
NIC1 and SPA-0 and SPB-0 are on one network subnet. NIC2 and SPA-1 and SPB-1 are
on a different subnet. This example connects NIC1 to SPA-0 and SPB-0, and NIC2 to
SPA-1 and SPB-1.
Note: These could also be on the same subnet, but we do not recommend it.
◆
NIC1
• SPA-0
• SPB-0
◆
NIC2
• SPA-1
• SPB-1
To configure the iSCSI Initiator manually, complete the following steps:
1.
While logged in as an Administrator on the server, open the Microsoft iSCSI
Initiator through Control Panel (showing All Control Panel Items) or
Administrative Tools.
Note: Do not use Quick Connect on the Targets tab. (If you used Quick Connect,
see “Windows 2008 R2 iSCSI Initiator cleanup” on page 35).
Windows 2008 R2 iSCSI Initiator manual procedure
31
iSCSI Attach Environments
2.
Select Discovery > Discover Portal, in the iSCSI Initiator Properties window:
The Discover Target Portal dialog box displays:
3.
32
Enter the IP Address of the Target Storage address and select Advanced.
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
The Advanced Setting dialog box displays:
4. Select Microsoft iSCSI Initiator in the Local adapter field.
5. Select the IP address of the NIC to be used.
6. Click OK and then OK again.
The iSCSI Initiator Properties window displays:
Windows 2008 R2 iSCSI Initiator manual procedure
33
iSCSI Attach Environments
7. Select the Targets tab.
8.
Highlight the first target iqn and select Connect.
The Connect to Target dialog box displays:
9. Select Enable multi-path if using Dell EMC PowerPath™ or Windows 2008 Native
MPIO.
10. Click Advanced.
The Advanced Settings dialog box displays:
11. In the Local adapter field, select Microsoft iSCSI Initiator from the drop-down
menu.
12. In the Initiator IP field, select the correct NIC IP address from the drop-down
menu.
13. In the Target poral IP field, select the IP address from the drop-down menu.
14. Click OK and then OK again.
15. Connect each of the other three targets in the list following the same procedure
listed in the previous steps.
34
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
16. In the iSCSI Initiator Properties window, select the Favorite Targets tab. This
should show each of the targets that have been connected:
17. If the host has Unisphere/Navisphere Agent installed, you should now see it logged
in and registered in Unisphere/Navisphere Manager. Otherwise you will need to
manually register the NIC in Unisphere/Navisphere Manager.
18. Place the host in a Storage Group that has LUNs in it using Unisphere/Navisphere
Manager, and then go back to the host and do a device manager Scan for
Hardware Changes. After a few minutes, you should see the disk devices arrive in
the PowerPath GUI and or in Disk Management.
Note: PowerPath only shows the one adapter in the PowerPath GUI, even though
you might be using multiple NICs. The adapter seen here does not represent the
NICs you have installed in your system, but rather it represents the MS iSCSI
software initiator.
Windows 2008 R2 iSCSI Initiator cleanup
Note: If running Windows 2008 Failover Cluster, ensure that this host does not own
any disk resources. Move resources to another node in the cluster or take the disk
resources offline.
Similarly, any LUNs being used on a Standalone Windows host need to be offline. Use
Disk Management to offline the disks.
Windows 2008 R2 iSCSI Initiator manual procedure
35
iSCSI Attach Environments
To clean up the iSCSI Initiator, complete the following steps:
1.
While logged in as an Administrator on the server, open the Microsoft iSCSI
Initiator through Control Panel (showing All Control Panel Items) or
Administrative Tools.
2. Select the Discovery tab, select one of the addresses in the Target Portals field,
and click Remove:
3.
A warning appears. Click OK:
4.
Remove all the other Target Portals.
5. In the iSCSI Initiator Properties window, select the Volumes and Devices tab,
select the volume in the Volume List field, and click Remove. Do this for each
volume you want to remove.
6. In the iSCSI Initiator Properties window, select the Favorite Targets tab, select
the target from the Favorite Targets field, and click Remove. Do this for each
target that you want to remove:
36
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
7. In the iSCSI Initiator Properties window, select the Targets tab, select one of the
targets in the Discovered targets field, and click Disconnect.
8. A warning message displays. Click Yes:
9. Follow steps Step 7 and Step 8 for each of the targets to be disconnected.
If you are running PowerPath, all of the devices will show as dead in the PowerPath
GUI. To clean and remove these, complete the following steps:
1. Open a command prompt using Run as administrator:
2. Type powermt check and when asked to remove dead device, select "a" for ALL.
3.
Check the Discovery tab to ensure that there are no further targets connected.
4.
Check each of the iSCSI initiator tabs and ensure they are all empty.
Windows 2008 R2 iSCSI Initiator manual procedure
37
iSCSI Attach Environments
Using MS iSNS server software with iSCSI configurations
The Microsoft iSNS Server is a Microsoft Windows service that processes iSNS
registrations, deregistrations, and queries via TCP/IP from iSNS clients, and also
maintains a database of these registrations. The Microsoft iSNS Server package
consists of Windows service software, a control-panel applet, a command-line
interface tool, and a set of WMI interfaces. Additionally, there are DLLs allowing
Microsoft Cluster Server to manage Microsoft iSNS Server as a cluster resource.
When configured properly, the iSNS server allows iSCSI initiators to query for available
iSCSI targets that are registered with the iSNS server. The iSNS server also allows
administration of iSCSI networks by providing a form of “zoning” in order to allow
initiators access only to targets designated by the administrator.
Prior to running the installation, we recommend that your iSCSI network interface
controller (NIC) be configured to work with your iSCSI network. Symmetrix, VNX
series, and CLARiiON iSCSI interfaces must be configured to recognize and register
with the iSNS server software. Refer to the DMX MPCD For iSCSI Version 1.0.0
Technical Notes for information on the configuration of the MPCD for iSCSI into
Symmetrix DMX systems. VNX series and CLARiiON configuration is done by an Dell
EMC Customer Engineer (CE) through Unisphere/Navisphere Manager. The CE will
configure your CX-Series system settings for each iSCSI port.
After your storage array target ports are configured, install the iSNS server software
by starting the installation package downloaded from the Microsoft website.
To install the iSNS server software, refer to Microsoft documentation at
http://microsoft.com.
38
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
iSCSI Boot with the Intel PRO/1000 family of adapters
Intel iSCSI Boot is designed for the Intel PRO/1000 family of PCI-Express Server
Adapters. Intel iSCSI Boot provides the capability to boot from a remote iSCSI disk
volume located on an iSCSI-based Storage Area Network (SAN).
The basic steps to configuring boot from SAN are:
1.
Prepare you storage array for boot from SAN.
2. Install boot-capable hardware in your system.
3. Install the latest Intel iSCSI Boot firmware using the iscsicli DOS utility.
4. Connect the host to a network with the contains the iSCSI target.
5. Configure the iSCSI boot firmware on the NIC to boot from a pre-configured iSCSI
target disk.
6. Configure the host to boot from the iSCSI target.
This section focuses on preparing your array to boot from SAN. The steps listed above
are documented on the Intel iSCSI Remote Boot Support page, with a list of supported
adapters.
Preparing your storage array for boot
This section explains how to prepare your array in order to successfully present a boot
LUN to you host.
The first thing you need to consider is what the host name will be. Using the naming
conventions explained in “Node-names” on page 30, record an appropriate iqn name
for your host.
In the following example, we will use an iqn name of
iqn.1992-05.com.microsoft:intel.hctlab.hct.
Configuring your CX3 for iSCSI boot
Note: The following example assumes that you are familiar with Unisphere/Navisphere
Manager.
For a boot from SAN to work properly, you first need to present a LUN to the host.
To do this on a CX3 using Navisphere Manager:
1. Create new initiator records that identify the host to the array.
2. Create a record for each SP port that you might potentially connect.
3. Create a storage group with the new server and boot LUN. This LUN should be
sized properly in order for the OS, and any other applications, to fit properly.
iSCSI Boot with the Intel PRO/1000 family of adapters
39
iSCSI Attach Environments
After this is complete, follow these steps:
1.
Right-click on the Storage Array and then select Connectivity Status:
2. Click New to display the Create Initiator Record window:
3. In the Create Initiator Record window:
a. Enter the iqn name of the host in the Initiator Name field. For our example, use
iqn.1992-05.com.microsoft:intel.hctlab.hct.
b. Select the SP - port to which the host will connect.
c. Enter a host name and ip address in the Host Information section.
d. Click OK.
40
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Once these steps are complete, the host displays in the Connectivity Status
window:
4. Remember that the above steps create an initiator record that has a path to B0. To
create additional initiator records, repeat steps 2 and 3 on page page 40.
a. Select a different SP - port in the drop-down menu.
b. Instead of entering a new host name, select Existing Host and choose the host
created during the creation of the initial initiator record.
Note: This is critical to provide uninterrupted access if an array side path failure
should occur.
5. Once you have created initiator records for each SP port, you will be able to create
a storage group as you normally would and assign the newly created host and a
boot LUN.
At this point you have configured the array to present a boot LUN to the host at boot
up. You can continue with the instructions documented by Intel to install your OS to a
local hard drive and then image your host OS to the boot LUN assigned above.
Post installation information
This section contains the following installation:
Using two Intel NICs in
a single host
◆
“Using two Intel NICs in a single host” on page 41
◆
“PowerPath for Windows” on page 42
The process in “Configuring your CX3 for iSCSI boot”, beginning on page 39, creates
connections that can be accessed by the host on all ports available on the CX3 (in our
example ports A0, A1, B0, and B1).
By using a dual-port PRO/1000, or two single-port PRO/1000s, the Intel BIOS will
allow you to set up one port as the primary and another as the secondary. By
configuring the primary login to connect to one SP, and the secondary login to connect
to the other SP, your host will have access to both SPs.
iSCSI Boot with the Intel PRO/1000 family of adapters
41
iSCSI Attach Environments
Note: You do not need to configure Microsoft iSCSI Software Initiator for Windows to
be able to detect the iSCSI Disk. Microsoft iSCSI Software Initiator automatically
retrieves the iSCSI configurations from the PRO/1000 adapter iSCSI Boot firmware.
PowerPath for
Windows
After you are successfully booting from your array, you can install PowerPath.
Before installing PowerPath ensure that your paths between the NICs and the array
ports are set up properly. To do this, use a combination of the Intel BIOS logins,
documented in the Intel guide in the "Firmware Setup" section, along with additional
target logins using the Microsoft Initiator. What you are trying to accomplish is a path
setup that looks much like what is discussed in “Using the Initiator with PowerPath” on
page 61.
When complete, your paths will look like those shown in Figure 1 on page 42.
Figure 1 Four paths
Setting this up will be slightly different than what is discussed in “Using the Initiator
with PowerPath” on page 61. Remember that you have already created two paths by
configuring the Intel BIOS with primary and secondary logins. So, for example, if you
configure the Intel BIOS to connect to A0 and B1, after you boot your host the
Microsoft Initiator will show two connected logins to port A0 and B1 on the Target tab.
To complete the path setup you need to use the process beginning with Step 1 on
page 62. When complete, you will have something similar to the following:
42
◆
NIC1 A0
◆
NIC2  A1
◆
NIC1B0
◆
NIC2  B1
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
You can now install PowerPath 4.6.x. Once installed, PowerPath Administrator will look
similar to Figure 2. Failures on the array side (loss of path between SP port and switch
and SP failures) will be managed correctly by PowerPath.
Figure 2 PowerPathAdmin
Note: PowerPath will sometimes show behavior that is not typically seen in non-boot
implementations because of the design of the boot version of the Microsoft Initiator.
The most notable difference is when a host side cable/NIC fault occurs. If the cable
connected to the NIC that first found the LUN at boot time is disconnected, or if the
NIC fails, PowerPath will show three dead paths instead of the two that would be
expected. This behavior is expected with the Microsoft Initiator boot version. If the
paths were set up as previously explained, a host side fault will not affect your system.
iSCSI Boot with the Intel PRO/1000 family of adapters
43
iSCSI Attach Environments
Notes on Microsoft iSCSI Initiator
This section contains important information about Microsoft iSCSI Initiator, including:
◆
“Microsoft Cluster Server” on page 60
◆
“Boot” on page 60
◆
“Boot” on page 60
◆
“NIC teaming” on page 61
◆
“Using the Initiator with PowerPath” on page 61
◆
“Commonly seen issues” on page 65
iSCSI failover behavior with the Microsoft iSCSI initiator
When creating an iSCSI session using the Microsoft iSCSI Initiator, you must choose in
the Advanced Settings dialog box (Figure 3 on page 44) is whether to:
◆
Have iSCSI traffic for that session travel over a specific NIC, or
◆
Allow the OS to choose which NIC will issue the iSCSI traffic.
This option also allows the Microsoft iSCSI Initiator to perform some failover
(independent of PowerPath) in the case of a NIC failure.
Note: Multiple subnet configurations are highly recommended as issues can arise in
single subnet configurations.
Figure 3 Advanced Settings dialog box
44
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
The Source IP pull-down menu in the Advanced Settings dialog box lists the IP
address of each NIC on the server, as well as an entry labeled Default. Default allows
Windows to choose which NIC to use for iSCSI traffic.
The examples in this section describe the different failover behaviors that can occur
when a NIC fails in both a single subnet configuration and a multiple subnet
configuration after choosing either a specific NIC from the pull-down menu or Default:
◆
“Single subnet, Source IP is "Default"” on page 47
◆
“Single subnet, Source IPs use specific NIC IP addresses” on page 49
◆
“Multiple subnets, Source IP is "Default"” on page 53
◆
“Multiple subnets, Source IPs use specific NIC IP addresses” on page 55
Single iSCSI subnet configuration
Figure 4 illustrates a single iSCSI subnet configuration.
Figure 4 Single iSCSI subnet configuration
In this configuration, there is a single subnet used for all iSCSI traffic. This iSCSI
subnet is routable with the corporate network, but only iSCSI ports on the array and
server NICs sending/receiving iSCSI traffic are connected to switches on this subnet.
The Windows server has a total of three NICs:
◆
One connected to the corporate network, with a defined default gateway
◆
Two connected to the iSCSI subnet, for NIC redundancy
Partial output from an ipconfig /all command from the server returns:
Ethernet adapter NIC1:
Description . .
IP Address. . .
Subnet Mask . .
Default Gateway
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. : Intel(R) PRO/1000 MT Dual Port Server Adapter
: 10.14.108.78
: 255.255.255.0
:
Ethernet adapter NIC2:
Description .
IP Address. . .
Subnet Mask . .
Default Gateway
. . . . . : Intel(R) PRO/1000 MT Dual Port Server Adapter #2
. . . . : 10.14.108.79
. . . . : 255.255.255.0
. . . . :
Notes on Microsoft iSCSI Initiator
45
iSCSI Attach Environments
Ethernet adapter Corporate:
Description . . . . . ..
IP Address. . . . . . . :
Subnet Mask . . . . . . :
Default Gateway . . . . :
. : Intel 8255x-based PCI Ethernet Adapter (10/100)
10.14.16.172
255.255.255.0
10.14.16.1
All four iSCSI ports on the VNX series and CLARiiON are also connected to the iSCSI
subnet. Each iSCSI port has a default gateway configured (with an iSCSI subnet
address). The management port on the VNX series and CLARiiON is connected to the
corporate network, and also has a default gateway defined.
The VNX series and CLARiiON's network configuration is as follows:
Management
iSCSI Port
iSCSI Port
iSCSI Port
iSCSI Port
port (10/100 Mb):IP Address 10.14.16.46,
SP A0: IP Address 10.14.108.46, default
SP A1: IP Address 10.14.108.48, default
SP B0: IP Address 10.14.108.47, default
SP B1: IP Address 10.14.108.49, default
default
gateway
gateway
gateway
gateway
gateway 10.14.16.1
10.14.108.1
10.14.108.1
10.14.108.1
10.14.108.1
Fully licensed PowerPath is installed for all examples.
ipconfig information
The corporate network is the 10.14.16 subnet. The Server's Intel 8255x-based PCI
Ethernet NIC connects to this subnet.
The iSCSI subnet is the 10.14.108 subnet. The Server's Intel Pro/1000 MT Dual Port
NICs connect to this subnet.
An ipconfig /all command from the server returns:
Windows IP Configuration
Host Name . . . . . . .
Primary Dns Suffix . .
Node Type . . . . . . .
IP Routing Enabled. . .
WINS Proxy Enabled. . .
DNS Suffix Search List.
.
.
.
.
.
: compaq8502
: YAMAHA.com
: Unknown
: No
: No
. : YAMAHA.com
Ethernet adapter NIC1:
Connection-specific DNS Suffix . :
Description . . . . . . . . . . . : Intel(R) PRO/1000 MT Dual Port Server Adapter
Address. . . . . .: 00-04-23-AB-83-42
DHCP Enabled. . . . . . . : No
IP Address. . . . . . . . : 10.14.108.78
Subnet Mask . . . . . . . : 255.255.255.0
Default Gateway . . . . . :
Ethernet adapter NIC2:
Connection-specific
Description . . . .
Physical Address. .
DHCP Enabled. . . .
IP Address. . . . .
Subnet Mask . . . .
Default Gateway . .
DNS
. .
. .
. .
. .
. .
. .
Physical
Suffix . :
. . . . . : Intel(R) PRO/1000 MT Dual Port Server Adapter #2
. .: 00-04-23-AB-83-43
. .: No
. .: 10.14.108.79
. .: 255.255.255.0
. .:
Ethernet adapter Corporate:
Connection-specific DNS Suffix . :
Description . . . . . . . . . . . : Intel 8255x-based PCI Ethernet Adapter (10/100)
Physical Address. . . . . . : 08-00-09-DC-E3-9C
DHCP Enabled. . . . . . . . : No
IP Address. . . . . . . . . : 10.14.16.172
Subnet Mask . . . . . . . . : 255.255.255.0
46
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Default Gateway . . . . .. : 10.14.16.1
DNS Servers . . . . . . . . : 10.14.36.200
10.14.22.13
Routing table information
A route print command from the server returns:
IPv4 Route Table
===========================================================================
Interface List
0x1 .......................... MS TCP Loopback interface
0x10003 ...00 04 23 ab 83 42 .....Intel(R) PRO/1000 MT Dual Port Server Adapter
0x10004 ...00 04 23 ab 83 43 ...... Intel(R) PRO/1000 MT Dual Port Server Adapter #2
0x10005 ...08 00 09 dc e3 9c ...... Intel 8255x-based PCI Ethernet Adapter (10/100)
===========================================================================
===========================================================================
Active Routes:
Network Destination
Netmask
Gateway
Interface Metric
0.0.0.0
0.0.0.0
10.14.16.1
10.14.16.172
20
10.14.16.0
255.255.255.0
10.14.16.172
10.14.16.172
20
10.14.16.172 255.255.255.255
127.0.0.1
127.0.0.1
20
10.14.108.0
255.255.255.0
10.14.108.78
10.14.108.78
10
10.14.108.0
255.255.255.0
10.14.108.79
10.14.108.79
10
10.14.108.78 255.255.255.255
127.0.0.1
127.0.0.1
10
10.14.108.79 255.255.255.255
127.0.0.1
127.0.0.1
10
10.255.255.255 255.255.255.255
10.14.16.172
10.14.16.172
20
10.255.255.255 255.255.255.255
10.14.108.78
10.14.108.78
10
10.255.255.255 255.255.255.255
10.14.108.79
10.14.108.79
10
127.0.0.0
255.0.0.0
127.0.0.1
127.0.0.1
1
224.0.0.0
240.0.0.0
10.14.16.172
10.14.16.172
20
224.0.0.0
240.0.0.0
10.14.108.78
10.14.108.78
10
224.0.0.0
240.0.0.0
10.14.108.79
10.14.108.79
10
255.255.255.255 255.255.255.255
10.14.16.172
10.14.16.172
1
255.255.255.255 255.255.255.255
10.14.108.78
10.14.108.78
1
255.255.255.255 255.255.255.255
10.14.108.79
10.14.108.79
1
Default Gateway:
10.14.16.1
===========================================================================
Persistent Routes:
None
Example 1 Single subnet, Source IP is "Default"
The Default setting can be verified through the iscsicli sessionlist command. The
bold output shows an Initiator Portal of 0.0.0.0/<TCP port>, which is what Default is
displayed as follows:
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
ffffffff8ae2600c-4000013700000002
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a0
40 00 01 37 00 00
1f 34
1
:
:
:
:
ffffffff8ae2600c-1
0.0.0.0/1049
10.14.108.46/3260
01 00
Notes on Microsoft iSCSI Initiator
47
iSCSI Attach Environments
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
ffffffff8ae2600c-4000013700000003
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8ae2600c-2
0.0.0.0/1050
10.14.108.48/3260
01 00
ffffffff8ae2600c-4000013700000004
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b0
40 00 01 37 00 00
5a 34
1
:
:
:
:
ffffffff8ae2600c-3
0.0.0.0/1051
10.14.108.47/3260
01 00
ffffffff8ae2600c-4000013700000005
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8ae2600c-4
0.0.0.0/1052
10.14.108.49/3260
01 00
The operation completed successfully.
When two NICs are on the same iSCSI traffic subnet (10.14.108), Windows will only use
one NIC for transmitting all iSCSI traffic. In this example, it uses the NIC1 (IP address
10.14.108.78), which is the highest entry in the routing table for the 10.14.108 subnet.
NIC1 handles iSCSI traffic to all four VNX series and CLARiiON iSCSI SP ports, while
NIC2 (10.14.108.79) is idle.
If NIC1 fails, Windows automatically fails over to NIC2 for all iSCSI traffic. This failure
is transparent to PowerPath, as shown in the following powermt display output:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=CLAROpt; priority=0; queued-IOs=2
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active alive
0
0
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
active alive
0
0
48
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
7 port7\path0\tgt2\lun1
7 port7\path0\tgt3\lun1
c7t2d1
c7t3d1
SP B0
SP B1
active
active
alive
alive
1
1
0
0
All paths remain listed as active no errors are indicated, and the Q-IOs Stats indicate
that both paths to SP B are still being used for traffic to this LUN.
If NIC1 is subsequently repaired, Windows does not return the iSCSI traffic to NIC1.
NIC1 remains idle while NIC2 is used for all iSCSI traffic to all four VNX series and
CLARiiON iSCSI SP ports. After NIC1 is repaired, it would take an NIC2 failure to move
iSCSI traffic back to NIC1.
Example 2 Single subnet, Source IPs use specific NIC IP addresses
The following iscsicli sessionlist output shows that each NIC is used for two iSCSI
sessions; four in total. The bold output below shows the IP addresses used in the
Initiator Portals.
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
:
Initiator Node Name
:
Target Node Name
:
Target Name
:
ISID
:
TSID
:
Number Connections
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
ffffffff8ae2700c-4000013700000002
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a0
40 00 01 37 00 00
5a 34
1
:
:
:
:
ffffffff8ae2700c-1
10.14.108.78/1394
10.14.108.46/3260
01 00
ffffffff8ae2700c-4000013700000003
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a1
40 00 01 37 00 00
df 84
1
:
:
:
:
ffffffff8ae2700c-2
10.14.108.79/1395
10.14.108.48/3260
01 00
ffffffff8ae2700c-4000013700000004
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b0
40 00 01 37 00 00
1f 34
1
:
:
:
:
ffffffff8ae2700c-3
10.14.108.78/1396
10.14.108.47/3260
01 00
: ffffffff8ae2700c-4000013700000005
: iqn.1991-05.com.microsoft:compaq8502.yamaha.com
Notes on Microsoft iSCSI Initiator
49
iSCSI Attach Environments
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b1
40 00 01 37 00 00
df 84
1
:
:
:
:
ffffffff8ae2700c-4
10.14.108.79/1397
10.14.108.49/3260
01 00
The operation completed successfully.
In this configuration both NICs are used for iSCSI traffic, even though they are on the
same subnet. iSCSI traffic targeted to VNX series and CLARiiON iSCSI SP ports A0
and B0 are directed through NIC1. iSCSI traffic targeted to VNX series and CLARiiON
iSCSI SP ports A1 and B1 are directed through NIC2.
If NIC1 fails, Windows will not attempt to re-route iSCSI traffic targeted to VNX series
and CLARiiON iSCSI SP ports A0 and B0, even though NIC2 can physically reach
those ports. Instead, Windows will fail iSCSI sessions connected to SP ports A0 and
B0, which in turn leads PowerPath to mark paths to those ports as “dead.” The
following powermt display output shows paths to A0 and B0 marked as dead. All
iSCSI traffic for this LUN is directed to the single surviving path on SP B, the current
owner on this LUN.
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=CLAROpt; priority=0; queued-IOs=2
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active dead
0
1
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
active alive
0
0
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active dead
0
1
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
active alive
2
0
If NIC1 is subsequently repaired, Windows re-establishes iSCSI sessions to VNX series
and CLARiiON SP ports A0 and B0, and PowerPath marks those paths as "alive" and
once again uses the path to B1 (as well as the path to B0) for IO to this LUN.
50
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Multiple iSCSI subnet configuration
Figure 5 illustrates a multiple iSCSI subnet configuration.
Figure 5 Multiple iSCSI subnet configuration
In this configuration, there are two subnets used for all iSCSI traffic:
◆
iSCSI Subnet 108 is the 10.14.108 subnet.
◆
iSCSI Subnet 109 is the 10.14.109 subnet.
These iSCSI subnets are routable with the corporate network and with each other, but
only iSCSI ports on the array and server NICs sending/receiving iSCSI traffic are
connected to switches on this subnet.
The Windows server has a total of three NICs:
◆
One connected to the corporate network, with a defined default gateway
◆
One connected to each of the two iSCSI subnets
Partial output from an ipconfig /all command from the server returns:
Ethernet adapter iSCSI108:
Description . . . . . .
IP Address. . . . . . .
Subnet Mask . . . . . .
Default Gateway . . . .
.
.
.
.
.
.
.
.
Ethernet adapter iSCSI109:
Description . . . . . .
IP Address. . . . . . .
Subnet Mask . . . . . .
Default Gateway . . . .
.
.
.
.
. . . : Intel(R) PRO/1000 MT Dual Port Server Adapter #2
: 10.14.109.78
: 255.255.255.0
:
Ethernet adapter Corporate:
Description . . . . . . .
IP Address. . . . . . . .
Subnet Mask . . . . . . .
Default Gateway . . . . .
.
:
:
:
. . . : Intel(R) PRO/1000 MT Dual Port Server Adapter
: 10.14.108.78
: 255.255.255.0
:
.. . : Intel 8255x-based PCI Ethernet Adapter (10/100)
10.14.16.172
255.255.255.0
10.14.16.1
Notes on Microsoft iSCSI Initiator
51
iSCSI Attach Environments
An iSCSI port on each Storage Processor of the VNX series and CLARiiON is
connected to each iSCSI subnet. This is done to create high availability in case of
either a subnet failure or an SP failure.
If a subnet fails, the remaining iSCSI subnet can continue to send iSCSI traffic to each
SP, allowing both SPs to continue to process IO, thereby spreading the IO load.
If an SP fails, PowerPath must trespass all LUNs to the surviving SP, but PowerPath is
able to load balance across both subnets, spreading the network load across them.
Each iSCSI port has a default gateway configured (with an iSCSI subnet address). The
management port on the VNX series and CLARiiON is connected to the corporate
network and also has a defined default gateway.
The VNX series and CLARiiON's network configuration is as follows:
Management
iSCSI Port
iSCSI Port
iSCSI Port
iSCSI Port
port (10/100 Mb):IP Address 10.14.16.46,
SP A0: IP Address 10.14.108.46, default
SP A1: IP Address 10.14.109.48, default
SP B0: IP Address 10.14.108.47, default
SP B1: IP Address 10.14.109.49, default
default
gateway
gateway
gateway
gateway
gateway 10.14.16.1
10.14.108.1
10.14.109.1
10.14.108.1
10.14.109.1
Fully licensed PowerPath is installed for all examples.
ipconfig information
The corporate network is the 10.14.16 subnet. The Server's Intel 8255x-based PCI
Ethernet NIC connects to this subnet.
The iSCSI subnets are the 10.14.108 and the 10.14.109 subnets. The Server's Intel
Pro/1000 MT Dual Port NICs connect to these subnets.
An ipconfig /all command from the server returns:
Windows IP Configuration
Host Name . . . . . . .
Primary Dns Suffix . .
Node Type . . . . . . .
IP Routing Enabled. . .
WINS Proxy Enabled. . .
DNS Suffix Search List.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Ethernet adapter iSCSI108:
Connection-specific DNS Suffix
Description . . . . . . . . . .
Physical Address. . . . . . . .
DHCP Enabled. . . . . . . . . .
IP Address. . . . . . . . . . .
Subnet Mask . . . . . . . . . .
Default Gateway . . . . . . . .
Ethernet adapter iSCSI109:
Connection-specific DNS Suffix
Description . . . . . . . . . .
Physical Address. . . . . . . .
DHCP Enabled. . . . . . . . . .
IP Address. . . . . . . . . . .
Subnet Mask . . . . . . . . . .
Default Gateway . . . . . . . .
Ethernet adapter Corporate:
Connection-specific DNS Suffix
Description . . . . . . . . . .
Physical Address. . . . . . . .
DHCP Enabled. . . . . . . . . .
IP Address. . . . . . . . . . .
52
.
.
.
.
.
.
:
:
:
:
:
:
compaq8502
YAMAHA.com
Unknown
No
No
YAMAHA.com
.
.
.
.
.
.
.
:
:
:
:
:
:
:
Intel(R) PRO/1000 MT Dual Port Server Adapter
00-04-23-AB-83-42
No
10.14.108.78
255.255.255.0
.
.
.
.
.
.
.
:
:
:
:
:
:
:
Intel(R) PRO/1000 MT Dual Port Server Adapter #2
00-04-23-AB-83-43
No
10.14.109.78
255.255.255.0
.
.
.
.
.
:
:
:
:
:
Intel 8255x-based PCI Ethernet Adapter (10/100)
08-00-09-DC-E3-9C
No
10.14.16.172
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . : 10.14.16.1
DNS Servers . . . . . . . . . . . : 10.14.36.200
10.14.22.13
Routing table information
A route print command from the server returns:
IPv4 Route Table
===========================================================================
Interface List
0x1 ........................... MS TCP Loopback interface
0x10003 ...00 04 23 ab 83 42 ...... Intel(R) PRO/1000 MT Dual Port Server Adapter
0x10004 ...00 04 23 ab 83 43 ...... Intel(R) PRO/1000 MT Dual Port Server Adapter #2
0x10005 ...08 00 09 dc e3 9c ...... Intel 8255x-based PCI Ethernet Adapter (10/100)
===========================================================================
===========================================================================
Active Routes:
Network Destination
Netmask
Gateway
Interface Metric
0.0.0.0
0.0.0.0
10.14.16.1
10.14.16.172
20
10.14.16.0
255.255.255.0
10.14.16.172
10.14.16.172
20
10.14.16.172 255.255.255.255
127.0.0.1
127.0.0.1
20
10.14.108.0
255.255.255.0
10.14.108.78
10.14.108.78
10
10.14.108.78 255.255.255.255
127.0.0.1
127.0.0.1
10
10.14.109.0
255.255.255.0
10.14.109.78
10.14.109.78
10
10.14.109.78 255.255.255.255
127.0.0.1
127.0.0.1
10
10.255.255.255 255.255.255.255
10.14.16.172
10.14.16.172
20
10.255.255.255 255.255.255.255
10.14.108.78
10.14.108.78
10
10.255.255.255 255.255.255.255
10.14.109.78
10.14.109.78
10
127.0.0.0
255.0.0.0
127.0.0.1
127.0.0.1
1
224.0.0.0
240.0.0.0
10.14.16.172
10.14.16.172
20
224.0.0.0
240.0.0.0
10.14.108.78
10.14.108.78
10
224.0.0.0
240.0.0.0
10.14.109.78
10.14.109.78
10
255.255.255.255 255.255.255.255
10.14.16.172
10.14.16.172
1
255.255.255.255 255.255.255.255
10.14.108.78
10.14.108.78
1
255.255.255.255 255.255.255.255
10.14.109.78
10.14.109.78
1
Default Gateway:
10.14.16.1
===========================================================================
Persistent Routes:
None
Example 3 Multiple subnets, Source IP is "Default"
The Default setting can be verified through the iscsicli sessionlist command. The
bold output shows an Initiator Portal of 0.0.0.0/<TCP port>:
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
ffffffff8b0aa904-4000013700000002
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a0
40 00 01 37 00 00
5a 34
1
:
:
:
:
ffffffff8b0aa904-1
0.0.0.0/1066
10.14.108.46/3260
01 00
: ffffffff8b0aa904-4000013700000003
Notes on Microsoft iSCSI Initiator
53
iSCSI Attach Environments
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8b0aa904-2
0.0.0.0/1067
10.14.109.48/3260
01 00
ffffffff8b0aa904-4000013700000004
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b0
40 00 01 37 00 00
5a 34
1
:
:
:
:
ffffffff8b0aa904-3
0.0.0.0/1068
10.14.108.47/3260
01 00
ffffffff8b0aa904-4000013700000005
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8b0aa904-4
0.0.0.0/1071
10.14.109.49/3260
01 00
The operation completed successfully.
In this configuration, NIC iSCSI108 will direct all iSCSI traffic targeted to VNX series
and CLARiiON SP ports A0 and B0 since they are all on the same subnet (10.14.108).
NIC iSCSI109 will direct all iSCSI traffic targeted to VNX series and CLARiiON SP
ports A1 and B1 since they are all on the same subnet (10.14.109).
If NIC iSCSI108 fails, Windows automatically fails over iSCSI traffic targeted to the
VNX series and CLARiiON SP ports on the 10.14.108 network. In this configuration, the
Corporate NIC (10.14.16.172) is chosen, since that NIC has a default gateway defined
and a routable network path to the 10.14.108 subnet.
This failure is transparent to PowerPath, as shown in the powermt display output
below:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=CLAROpt; priority=0; queued-IOs=2
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
54
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
7
7
7
7
port7\path0\tgt0\lun1
port7\path0\tgt1\lun1
port7\path0\tgt2\lun1
port7\path0\tgt3\lun1
c7t0d1
c7t1d1
c7t2d1
c7t3d1
SP
SP
SP
SP
A0
A1
B0
B1
active
active
active
active
alive
alive
alive
alive
0
0
1
1
0
0
0
0
All paths remain listed as active, no errors are indicated, and the Q-IOs Stats indicate
that both paths to SP B are still being used for traffic to this LUN. Traffic to SP port B0
is being routed through the 10.14.16 subnet to the 10.14.108 subnet, while traffic to SP
port B1 remains on the 10.14.109 subnet.
Since the Corporate network is now in use for iSCSI traffic, there may be performance
implications from this failover. If the Corporate network or Corporate NIC is at a slower
speed than the iSCSI subnet, throughput will be reduced. Additionally, iSCSI traffic is
now competing on the same wire with non-iSCSI traffic, which may cause network
congestion and/or reduced response times for both. If this is a concern, there are
multiple ways to avoid this issue:
◆
Configure your iSCSI subnets so they are not routable with other subnets or the
corporate network.
◆
Configure your iSCSI sessions to use a specific NIC, as described in “Multiple
subnets, Source IPs use specific NIC IP addresses” on page 55.
◆
Do not configure a default gateway on the VNX series and CLARiiON iSCSI ports.
This prevents iSCSI traffic on these ports from leaving their subnet.
If the Corporate NIC were to subsequently fail, Windows would not be able to failover
iSCSI traffic targeted to SP ports on the 10.14.108 subnet since no default gateway
exists for the lone-surviving NIC iSCSI109 on the 10.14.109 subnet. In this case,
PowerPath would mark paths to SP ports A0 and B0 as "dead."
Assuming only NIC iSCSI108 failed, when NIC iSCSI108 is subsequently repaired,
Windows does return the iSCSI traffic to NIC iSCSI108 since that is the shortest route
to the 10.14.108 subnet.
Example 4 Multiple subnets, Source IPs use specific NIC IP addresses
The following iscsicli sessionlist output shows that each iSCSI NIC is used for two
iSCSI sessions; four in total. The iSCSI sessions were created such that traffic directed
to a VNX series and CLARiiON SP port is routed through the NIC on that SP port's
subnet. The bold output shows the IP addresses used in the Initiator Portals:
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
ffffffff8ae2f424-4000013700000002
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a0
40 00 01 37 00 00
1f 34
1
:
:
:
:
ffffffff8ae2f424-1
10.14.108.78/1050
10.14.108.46/3260
01 00
Notes on Microsoft iSCSI Initiator
55
iSCSI Attach Environments
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
Session Id
Initiator Node Name
Target Node Name
Target Name
ISID
TSID
Number Connections
:
:
:
:
:
:
:
Connections:
Connection Id
Initiator Portal
Target Portal
CID
ffffffff8ae2f424-4000013700000003
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.a1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8ae2f424-2
10.14.109.78/1051
10.14.109.48/3260
01 00
ffffffff8ae2f424-4000013700000004
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b0
40 00 01 37 00 00
5a 34
1
:
:
:
:
ffffffff8ae2f424-3
10.14.108.78/1052
10.14.108.47/3260
01 00
ffffffff8ae2f424-4000013700000006
iqn.1991-05.com.microsoft:compaq8502.yamaha.com
(null)
iqn.1992-04.com.emc:cx.apm00063505574.b1
40 00 01 37 00 00
1a 85
1
:
:
:
:
ffffffff8ae2f424-5
10.14.109.78/1054
10.14.109.49/3260
01 00
The operation completed successfully.
In this configuration, NIC iSCSI108 will direct all iSCSI traffic targeted to VNX series
and CLARiiON SP ports A0 and B0 since they are all on the same subnet (10.14.108).
NIC iSCSI109 will direct all iSCSI traffic targeted to VNX series and CLARiiON SP
ports A1 and B1 since they are all on the same subnet (10.14.109).
If NIC iSCSI108 fails, Windows ON iSCSI SP ports A0 and B0, even though the
Corporate NIC can physically reach those ports through its default gateway. Instead,
Windows will fail iSCSI sessions connected to SP ports A0 and B0, which in turn leads
PowerPath to mark paths to those ports as "dead."
The following powermt display output shows paths to A0 and B0 marked as dead. All
iSCSI traffic for this LUN is directed to the single surviving path on SP B, the current
owner on this LUN.
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=CLAROpt; priority=0; queued-IOs=2
Owner: default=SP B, current=SP BArray failover mode: 1
56
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active dead
0
1
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
active alive
0
0
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active dead
0
1
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
active alive
2
0
If NIC iSCSI108 is subsequently repaired, Windows re-establishes iSCSI sessions to
VNX series and CLARiiON SP ports A0 and B0, and PowerPath marks those paths as
"alive" and once again uses the path to B1 (as well as the path to B0) for IO to this
LUN.
Unlicensed PowerPath and iSCSI failover behavior
Unlicensed PowerPath provides basic failover for hosts connected to VNX series and
CLARiiON systems. Unlicensed PowerPath will only use a single path from a host to
each SP for IO. Any paths besides these two paths are labeled “unlicensed” and
cannot be used for IO. The Microsoft iSCSI Initiator always logs in to its targets in a
specific order, connecting all SP A paths before connecting any SP B paths, and
always connecting to the ports in numeric order. For example, the MS iSCSI Initiator
will log in to port A0 first, then A1, A2, etc., for all SP A ports, then port B0, B1, etc.,
until the last SP B port. Unlicensed PowerPath chooses the first path discovered on
each SP for IO. In all of the examples in this section, PowerPath will only use paths to
A0 and B0 for IO.
The following is an example of a powermt display command with unlicensed
PowerPath:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy= BasicFailover; priority=0; queued-IOs=0
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active alive
0
0
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
active unlic
0
0
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active alive
0
0
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
active unlic
0
0
Unlicensed PowerPath does not change any of the failover behavior in Example 1 on
page 47 and Example 3 on page 53, where the Source IP address is “Default”. This is
because PowerPath does not know that a NIC has failed in these examples since the
MS iSCSI Initiator automatically fails over iSCSI sessions to a surviving NIC. The only
impact unlicensed PowerPath has in these examples is that only a single path will be
used for IO to an SP (to SP ports A0 and B0).
Unlicensed PowerPath will have an impact on failover behavior in Example 2 on
page 49 and Example 4 on page 55, where the Source IP address uses a specific NIC IP
address.
Revisiting Example 2 on page 49 with unlicensed PowerPath, the following shows an
abridged version of the iscsicli sessionlist command output:
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Notes on Microsoft iSCSI Initiator
57
iSCSI Attach Environments
Session Id
Target Name
: ffffffff8ae2700c-4000013700000002
: iqn.1992-04.com.emc:cx.apm00063505574.a0
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: ffffffff8ae2700c-4000013700000003
: iqn.1992-04.com.emc:cx.apm00063505574.a1
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: 10.14.108.79/1395
: 10.14.108.48/3260
: ffffffff8ae2700c-4000013700000004
: iqn.1992-04.com.emc:cx.apm00063505574.b0
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: 10.14.108.78/1394
: 10.14.108.46/3260
: 10.14.108.78/1396
: 10.14.108.47/3260
: ffffffff8ae2700c-4000013700000005
: iqn.1992-04.com.emc:cx.apm00063505574.b1
Connections:
Initiator Portal
Target Portal
: 10.14.108.79/1397
: 10.14.108.49/3260
The operation completed successfully.
If NIC1 fails (10.14.108.78), Windows will not attempt to re-route iSCSI traffic targeted
to VNX series and CLARiiON iSCSI ports A0 and B0, even though NIC2 (10.14.108.79)
can physically reach those ports. Instead, Windows will fail iSCSI sessions connected
to SP ports A0 and B0, which in turn leads PowerPath to mark those paths as dead.
However, the surviving two paths (to SP ports A1 and B1) are not licensed and not used
for IO, as shown in the following powermt display output:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=BasicFailover; priority=0; queued-IOs=0
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active dead
0
1
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
unlic
alive
0
0
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active dead
0
1
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
unlic
alive
0
0
Since all paths are either dead or unlicensed, IO fails because no usable path exists
from the host to the VNX series and CLARiiON.
However, a configuration can be designed that will avoid IO errors with a single NIC
failure and unlicensed PowerPath. Revisiting Example 4 on page 55 with unlicensed
PowerPath, the array IP addresses can be changed so that SP ports A0 and B0 do not
reside on the same subnet. The VNX series and CLARiiON’s network configuration
would be as follows:
58
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Management
iSCSI Port
iSCSI Port
iSCSI Port
iSCSI Port
port (10/100 Mb):
SP A0: IP Address
SP A1: IP Address
SP B0: IP Address
SP B1: IP Address
IP Address 10.14.16.46, default gateway 10.14.16.1
10.14.108.46, default gateway 10.14.108.1
10.14.109.48, default gateway 10.14.109.1
10.14.109.49, default gateway 10.14.109.1
10.14.108.47, default gateway 10.14.108.1
Using this array IP configuration, the following shows an abridged version of the
iscsicli sessionlist command output:
Microsoft iSCSI Initiator version 2.0 Build 1941
Total of 4 sessions
Session Id
Target Name
: ffffffff8a8d9204-4000013700000003
: iqn.1992-04.com.emc:cx.apm00063505574.a0
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: ffffffff8a8d9204-4000013700000004
: iqn.1992-04.com.emc:cx.apm00063505574.a1
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: 10.14.109.78/2788
: 10.14.109.48/3260
: ffffffff8a8d9204-4000013700000005
: iqn.1992-04.com.emc:cx.apm00063505574.b0
Connections:
Initiator Portal
Target Portal
Session Id
Target Name
: 10.14.108.78/2787
: 10.14.108.46/3260
: 10.14.109.78/2789
: 10.14.109.49/3260
: ffffffff8a8d9204-4000013700000006
: iqn.1992-04.com.emc:cx.apm00063505574.b1
Connections:
Initiator Portal
Target Portal
: 10.14.108.78/2790
: 10.14.108.47/3260
The operation completed successfully.
Output from a powermt display command shows that paths to VNX series and
CLARiiON SP ports A0 and B0 are usable for IO:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=BasicFailover; priority=0; queued-IOs=5
Owner: default=SP B, current=SP BArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active alive
0
0
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
unlic
alive
0
0
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active alive
5
0
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
unlic
alive
0
0
Notes on Microsoft iSCSI Initiator
59
iSCSI Attach Environments
In this configuration, NIC “iSCSI109” (10.14.109.78) is directing all iSCSI traffic to VNX
series and CLARiiON SP port B0 for this LUN, since this is the only path that is usable
by unlicensed PowerPath to this LUN’s owning SP. If NIC “iSCSI109” fails, Windows
will not attempt to re-route iSCSI traffic targeted to VNX series and CLARiiON iSCSI
SP port B0, even though the “Corporate” NIC can physically reach this port through
its default gateway. Instead, Windows will fail the iSCSI session connected to SP port
B0 (as well as the session connected to SP port A1), which in turn leads PowerPath to
mark paths to those SP ports as dead, as indicated in the following powermt display
output:
Pseudo name=harddisk2
CLARiiON ID=APM00063505574 [Compaq8502]
Logical device ID=600601609FD119003CE4D89460A6DB11 [LUN 44]
state=alive; policy=BasicFailover; priority=0; queued-IOs=5
Owner: default=SP B, current=SP AArray failover mode: 1
==============================================================================
---------------- Host --------------- Stor -- I/O Path - -- Stats --### HW Path
I/O Paths
Interf.
Mode
State Q-IOs Errors
==============================================================================
7 port7\path0\tgt0\lun1
c7t0d1
SP A0
active alive
5
0
7 port7\path0\tgt1\lun1
c7t1d1
SP A1
unlic
dead
0
1
7 port7\path0\tgt2\lun1
c7t2d1
SP B0
active dead
0
1
7 port7\path0\tgt3\lun1
c7t3d1
SP B1
unlic
alive
0
0
However, a usable path for unlicensed PowerPath to this LUN still exists – the path
from NIC “iSCSI108” (10.14.108.78) to VNX series and CLARiiON SP port A0.
Therefore, PowerPath will trespass the LUN to SP A and successfully direct IO to this
LUN through this surviving path.
If NIC “iSCSI109” is subsequently repaired, Windows will re-establish iSCSI sessions to
VNX series and CLARiiON SP ports B0 and A1, and unlicensed PowerPath will mark
these paths as alive. Additionally, since unlicensed PowerPath now has a healthy path
to this LUN’s default SP, PowerPath will auto-restore this LUN by trespassing it back
to SP B and then directing all iSCSI traffic to the path from NIC “iSCSI109” to VNX
series and CLARiiON SP port B0.
Microsoft Cluster Server
Microsoft Cluster Server (MSCS) shared storage (including the quorum disk) can be
implemented using iSCSI disk volumes as the shared storage. There is no special iSCSI
cluster or application configuration needed to support this scenario. Since the cluster
service manages application dependencies, it is not needed to make any cluster
managed service (or the cluster service itself) dependent upon the Microsoft iSCSI
service.
Microsoft MPIO and the Microsoft iSCSI DSM can be used with MSCS.
Boot
Currently, it is not possible to boot a Windows system using an iSCSI disk volume
provided by the Microsoft software iSCSI Initiator kernel mode driver. It is possible to
boot a Windows system using an iSCSI disk volume provided by an iSCSI HBA. The
only currently supported method for booting a Windows system using an iSCSI disk
volume is via a supported HBA.
60
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
NIC teaming
Microsoft does not support the use of NIC teaming on iSCSI interfaces. For iSCSI SAN
interfaces, Microsoft recommends that customers use dual or quad ported NICs or
multiple single-port NICs and allow Microsoft Windows components to handle
fail-over and redundancy to ensure consistent visibility into the data path. Fail-over and
load balancing of multiple paths to an iSCSI target from within the same server is
supported through Microsoft MPIO and multiple connections per session (fail over and
load balancing using multiple adapters in the same server.
Microsoft Cluster Server can also be used for fail over and load balancing of shared
storage resources between servers (fail over and load balancing between servers). NIC
teaming can still be used but only on LAN interfaces that are not used to connect to an
iSCSI SAN.
Using the Initiator with PowerPath
This section provides an overview of how to set up the Initiator for use with PowerPath
and describes how to use the Initiator to create the paths that PowerPath then takes
control of. This setup can be compared to creating zones in a Fibre Channel
environment.
Note: PowerPath version 5.1 SP2 is the minimum version required for support on
Windows Server 2008. Upon installing PowerPath, the installer will load the MPIO
feature and claim all disks associated with the Microsoft Initiator. During the install a
DOS windows will open. At this point, the MPIO feature is loaded. Do not close this
window. Once the MPIO feature is installed, PowerPath will close this window
automatically.
There are no manual steps that need to be done to configure MPIO. PowerPath will
perform all the required steps as part of the PowerPath install.
The Initiator allows you to log in multiple paths to the same target and aggregate the
duplicate devices into a single device exposed to Windows. Each path to the target can
be established using different NICs, network infrastructure, and target ports. If one
path fails then another session can continue processing I/O without interruption to the
application. It is PowerPath that aggregates these paths and manages them to provide
uninterrupted access to the device.
PowerPath uses the Microsoft MPIO framework that is installed with the Initiator in
conjunction with Dell EMC's DSM to provide multipathing functionality.
This section also describes how to use the Initiator to create the paths that PowerPath
then takes control of. This setup can be compared to creating zones in a fibre channel
environment.
This section is based on the following assumption:
Notes on Microsoft iSCSI Initiator
61
iSCSI Attach Environments
◆
The following two network cards are installed in the server:
• NIC1 (192.168.150.155)
• NIC2 (192.168.150.156)
◆
We will log into the target on four different ports:
• A0 (192.168.150.102)
• A1 (192.168.150.103)
• B0 (192.168.150.104)
• B1 (192.168.150.105).
Logging into a target.
◆
We will log into the target to create four separate paths:
• NIC1 A0
• NIC2  A1
• NIC1B0
• NIC2  B1
Follow the next steps to log in to the target.
1.
Select the port to log into and click Log On.
Figure 6 iSCSI Initiator Properties dialog box
62
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
The Log On to Target dialog box displays.
Figure 7 Log On to Target dialog box
2. Click Advanced.
The Advanced Settings dialog box displays.
Figure 8 Advanced Settings dialog box
Notes on Microsoft iSCSI Initiator
63
iSCSI Attach Environments
3. For the source IP choose the IP address associated with NIC1 and the Target Portal
(port) associated with port A0 on the target. Enter CHAP information, if needed,
and click OK.
To create the other paths to the array repeat the steps above for each path you want to
create.
In our example, you would choose a different port on the Targets tab (A1, B0….).
Note: You can only log into a port once from a host. Multiple logins are not supported.
On the Advanced Settings page choose your source and Target IPs.
Once completed, you should have four paths as shown in Figure 9.
Figure 9 Four paths
You can now install PowerPath, which will aggregate all four paths and present one
device to Windows Disk Management. From a host perspective, a failure to a NIC or
the path between and NIC and the target will be managed properly by PowerPath and
allow access to the device on the target. For example, if NIC1 fails the host would still
have an active paths to both SPs since NIC2 has a path to SPA1 and SPB1.
64
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Commonly seen issues
This section lists some of the common issues E-Lab has seen during testing. For a more
detailed list, refer to the Microsoft Initiator 2.0 Users Guide.
Multipathing
This section lists multipathing errors that may be recorded in the event logs and
discusses potential solutions for these error conditions. Even though Dell EMC does
not support multiple connections per session (MCS) on VNX series and CLARiiON and
Symmetrix, it is easy to confuse the configuration of MPIO and MCS.
Note: Dell EMC does support multiple connections per session (MCS) on VNXe. Refer
to Dell EMC knowledgebase solution, How to Configure Windows Multipathing on a
VNXe at http://support.emc.com/kb/14664, for more information.
The following errors might help point out that you are actually configuring MCS
instead of MPIO.
◆
Error: "Too many Connections" when you attempt to add a second connection to an
existing session.
This issue can occur if the target does not support multiple connections per
session (MCS). Consult with the target vendor to see if they plan on adding
support for MCS.
◆
When you attempt to add a second connection to an existing session, you may
notice that the Add button within the Session Connections window is grayed out.
This issue can occur if you logged onto the target using an iSCSI HBA that doesn't
support MCS.
For more information on MCS vs. MPIO, refer to the Microsoft iSCSI Software Initiator
2.x Users Guide located at http://microsoft.com.
Long boot time
If your system takes a long time to display the login prompt after booting, or it takes a
long time to log in after entering your login ID and password, then there may be an
issue related to the Microsoft iSCSI Initiator service starting. First see the "Running
automatic start services on iSCSI disks" section in the Microsoft Initiator 2.x Users
Guide for information about persistent volumes and the binding operation. Check the
system event log to see if there is an event Timeout waiting for iSCSI
persistently bound volumes… If this is the case, then one or more of the
persistently bound volumes did not reappear after reboot which could be due to
network or target error.
Another error you may see on machines that are slow to boot is an event log message
Initiator Service failed to respond in time to a request to
encrypt or decrypt data if you have persistent logins that are configured to use
CHAP. Additionally, the persistent login will fail to log in. This is due to a timing issue in
the service startup order. To work around this issue, increase the timeout value for the
IPSecConfigTimeout value in the registry under:
HKLM\SYSTEM\CurrentControlSet\Control\Class\{4D36E97B-E325-11CE-BFC1-080
02BE10318}\<Instance Number> \Parameters
Notes on Microsoft iSCSI Initiator
65
iSCSI Attach Environments
This has been seen in some cases where clusters are present.
Logging out of target
The MS iSCSI Initiator service will not allow a session to be logged out if there are any
handles that are open to the device. In this case, if the session is attempted to be
logged out, the error The session cannot be logged out since a device
on that session is currently being used. is reported. This means that
there is an application or device which has an open handle to the physical disk on the
target. If you look in the system event log you should see an event that has the name of
the device with the open handle.
Other event log errors of note
The source for system events logged by Software Initiator will be iScsiPrt. The
message in the log would convey the cause of that event.
Some of the common events are listed below. A complete list of events can be found in
the Microsoft Initiator User’s Guide 2.x.
Event ID 1
Initiator failed to connect to the target. Target IP address and TCP Port number
are given in dump data.
This event is logged when the Initiator could not make a TCP connection to the given
target portal. The dump data in this event will contain the target IP address and TCP
port to which Initiator could not make a TCP connection.
Event ID 9
Target did not respond in time for a SCSI request. The CDB is given in the dump
data.
This event is logged when the target did not complete a SCSI command within the
timeout period specified by SCSI layer. The dump data will contain the SCSI Opcode
corresponding to the SCSI command. User can refer to SCSI specification for getting
more information about the SCSI command.
Event ID 20
Connection to the target was lost. The Initiator will attempt to retry the
connection.
This event is logged when the Initiator loses connection to the target when the
connection was in iSCSI Full Feature Phase. This event typically happens when there
are network problems, a network cable is removed, a network switch is shutdown, or
target resets the connection. In all cases Initiator will attempt to reestablish the TCP
connection.
Event ID 34
A connection to the target was lost, but Initiator successfully reconnected to the
target. Dump data contains the target name.
This event is logged when the Initiator successfully reestablishes a TCP connection to
the target.
Troubleshooting
Note the following potential problems and their solutions.
Problem
66
Adding a Symmetrix IP address to the target portals returns initiator error.
Dell EMC Host Connectivity Guide for Windows
iSCSI Attach Environments
Solution
Problem
◆
Verify that the iSCSI login parameter is correct.
◆
Verify that the Volume Logix setup is correct.
Adding a Symmetrix IP address to the target portals returns authentication error.
Solution
Problem
◆
Verify that the CHAP feature is enabled, and that the user name and secret are
correct.
◆
Verify that the Symmetrix system has the correct user name and secret.
Login to the Symmetrix target returns The target had already been logged via a
Symmetrix.
Solution
Problem
◆
Press the Refresh button to verify that only one iSCSI session is established.
◆
Log out of the current iSCSI session to the Symmetrix system, and log in again.
Adding a target portal returns Connection Failed.
Solution
Problem
◆
Ensure that the IP address of the Symmetrix system is correct.
◆
Ensure the connectivity is correct by using the ping utility to the Symmetrix GE
port and vice versa.
File shares on iSCSI devices may not be re-created when you restart your computer.
Solution
This issue can occur when the iSCSI Initiator service is not initialized when the Server
service initializes. The Server service creates file shares. However, because iSCSI disk
devices are not available, the Server service cannot create file shares for iSCSI devices
until the iSCSI service is initialized. To resolve this issue, follow these steps on the
affected server:
1.
Make the Server service dependant on the iSCSI Initiator service.
2. Configure the BindPersistentVolumes option for the iSCSI Initiator service.
3. Configure persistent logons to the target. To do this, use one of the following
methods.
Method 1
a. Double-click iSCSI Initiator in Control Panel.
b. Click the Available Targets tab.
c. Click a target in the Select a target list, and then click Log On.
d. Click to select the Automatically restore this connection when the system
boots check box.
e. Click OK.
Notes on Microsoft iSCSI Initiator
67
iSCSI Attach Environments
Method 2
a. Click Start, click Run, type cmd, and then click OK.
b. At the command prompt, type the following command, and then press ENTER:
scsicli persistentlogintarget target_iqn T * * * * * * * * * * * * * * * 0
Note: target_iqn is the iSCSI qualified name (iqn) of the target.
Note: This resolution applies only when you specifically experience this issue with
the iSCSI Initiator service. Refer to Microsoft Knowledge Base article 870964 for
more information.
68
Dell EMC Host Connectivity Guide for Windows
CHAPTER 3
Invisible Body Tag
Virtual
Provisioning
This chapter provides information about Virtual Provisioning in a Windows
environment.
Note: For further information about the correct implementation of Virtual Provisioning,
refer to the Symmetrix Virtual Provisioning Implementation and Best Practices
Technical Note, available on Dell EMC Online Support
◆
◆
◆
Virtual Provisioning on Symmetrix.............................................. 70
Implementation considerations................................................... 74
Operating system characteristics............................................... 78
Virtual Provisioning
69
Virtual Provisioning
Virtual Provisioning on Symmetrix
Dell EMC Virtual Provisioning™ enables organizations to improve speed and ease of
use, enhance performance, and increase capacity utilization for certain applications
and workloads. Virtual Provisioning integrates with existing device management,
replication, and management tools, enabling customers to easily build Virtual
Provisioning into their existing storage management processes. Figure 10 shows an
example of Virtual Provisioning on Symmetrix.
Virtual Provisioning, which marks a significant advancement over technologies
commonly known in the industry as “thin provisioning,” adds a new dimension to tiered
storage in the array, without disrupting organizational processes.
Figure 10 Virtual Provisioning on Symmetrix
70
Dell EMC Host Connectivity Guide for Windows
Virtual Provisioning
Terminology
This section provides common terminology and definitions for Symmetrix and thin
provisioning.
Symmetrix
Thin provisioning
Basic Symmetrix terms include:
Device
A logical unit of storage defined within an
array.
Device capacity
The storage capacity of a device.
Device extent
Specifies a quantum of logically
contiguous blocks of storage.
Host accessible device
A device that can be made available for
host use.
Internal device
A device used for a Symmetrix internal
function that cannot be made accessible
to a host.
Storage pool
A collection of internal devices for some
specific purpose.
Basic thin provisioning terms include:
Thin device
A host accessible device that has no
storage directly associated with it.
Data device
An internal device that provides storage
capacity to be used by thin devices.
Thin pool
A collection of data devices that provide
storage capacity for thin devices.
Thin pool capacity
The sum of the capacities of the member
data devices.
Thin pool allocated capacity
A subset of thin pool enabled capacity that
has been allocated for the exclusive use of
all thin devices bound to that thin pool.
Thin device user
pre-allocated capacity
The initial amount of capacity that is
allocated when a thin device is bound to a
thin pool. This property is under user
control.
Bind
Refers to the act of associating one or
more thin devices with a thin pool.
Pre-provisioning
An approach sometimes used to reduce
the operational impact of provisioning
storage. The approach consists of
satisfying provisioning operations with
larger devices that needed initially, so that
the future cycles of the storage
provisioning process can be deferred or
avoided.
Virtual Provisioning on Symmetrix
71
Virtual Provisioning
Over-subscribed thin pool
A thin pool whose thin pool capacity is less
than the sum of the reported sizes of the
thin devices using the pool.
Thin device extent
The minimum quantum of storage that
must be mapped at a time to a thin device.
Data device extent
The minimum quantum of storage that is
allocated at a time when dedicating
storage from a thin pool for use with a
specific thin device.
Management tools
Configuring, replicating, managing, and monitoring thin devices and thin pools involve
the same tools and the same or similar functions as those used to manage traditional
arrays.
Use Symmetrix Management Console or Solutions Enabler to configure and manage
Virtual Provisioning.
Thin device
Symmetrix Virtual Provisioning introduces a new type of host-accessible device called
a thin device that can be used in many of the same ways that regular host-accessible
Symmetrix devices have traditionally been used. Unlike regular Symmetrix devices,
thin devices do not need to have physical storage completely allocated at the time the
devices are created and presented to a host. The physical storage that is used to
supply disk space for a thin device comes from a shared thin storage pool that has
been associated with the thin device.
A thin storage pool is comprised of a new type of internal Symmetrix device called a
data device that is dedicated to the purpose of providing the actual physical storage
used by thin devices. When they are first created, thin devices are not associated with
any particular thin pool. An operation referred to as binding must be performed to
associate a thin device with a thin pool.
When a write is performed to a portion of the thin device, the Symmetrix allocates a
minimum allotment of physical storage from the pool and maps that storage to a region
of the thin device, including the area targeted by the write. The storage allocation
operations are performed in small units of storage called data device extents. A
round-robin mechanism is used to balance the allocation of data device extents across
all of the data devices in the pool that have remaining unused capacity.
When a read is performed on a thin device, the data being read is retrieved from the
appropriate data device in the storage pool to which the thin device is bound. Reads
directed to an area of a thin device that has not been mapped does not trigger
allocation operations. The result of reading an unmapped block is that a block in which
each byte is equal to zero will be returned. When more storage is required to service
existing or future thin devices, data devices can be added to existing thin storage
pools. New thin devices can also be created and associated with existing thin pools.
It is possible for a thin device to be presented for host use before all of the reported
capacity of the device has been mapped. It is also possible for the sum of the reported
capacities of the thin devices using a given pool to exceed the available storage
capacity of the pool. Such a thin device configuration is said to be over-subscribed.
72
Dell EMC Host Connectivity Guide for Windows
Virtual Provisioning
Figure 11 Thin device and thin storage pool containing data devices
In Figure 11, as host writes to a thin device are serviced by the Symmetrix array, storage
is allocated to the thin device from the data devices in the associated storage pool. The
storage is allocated from the pool using a round-robin approach that tends to stripe the
data devices in the pool.
Virtual Provisioning on Symmetrix
73
Virtual Provisioning
Implementation considerations
When implementing Virtual Provisioning, it is important that realistic utilization
objectives are set. Generally, organizations should target no higher than 60 percent to
80 percent capacity utilization per pool. A buffer should be provided for unexpected
growth or a “runaway” application that consumes more physical capacity than was
originally planned for. There should be sufficient free space in the storage pool equal
to the capacity of the largest unallocated thin device.
Organizations also should balance growth against storage acquisition and installation
time frames. It is recommended that the storage pool be expanded before the last 20
percent of the storage pool is utilized to allow for adequate striping across the existing
data devices and the newly added data devices in the storage pool.
Thin devices can be deleted once they are unbound from the thin storage pool. When
thin devices are unbound, the space consumed by those thin devices on the associated
data devices is reclaimed.
Note: Users should first replicate the data elsewhere to ensure it remains available for
use.
Data devices can also be disabled and/or removed from a storage pool. Prior to
disabling a data device, all allocated tracks must be removed (by unbinding the
associated thin devices). This means that all thin devices in a pool must be unbound
before any data devices can be disabled.
The following information is provided in this section:
◆
“Over-subscribed thin pools” on page 74
◆
“Thin-hostile environments” on page 75
◆
“Pre-provisioning with thin devices in a thin hostile environment” on page 75
◆
“Host boot/root/swap/dump devices positioned on Symmetrix VP (tdev) devices”
on page 76
◆
“Cluster configurations” on page 76
Over-subscribed thin pools
It is permissible for the amount of storage mapped to a thin device to be less than the
reported size of the device. It is also permissible for the sum of the reported sizes of
the thin devices using a given thin pool to exceed the total capacity of the data devices
comprising the thin pool. In this case the thin pool is said to be over-subscribed.
Over-subscribing allows the organization to present larger-than-needed devices to
hosts and applications without having to purchase enough physical disks to fully
allocate all of the space represented by the thin devices.
The capacity utilization of over-subscribed pools must be monitored to determine
when space must be added to the thin pool to avoid out-of-space conditions.
Not all operating systems, filesystems, logical volume managers, multipathing
software, and application environments will be appropriate for use with
over-subscribed thin pools. If the application, or any part of the software stack
underlying the application, has a tendency to produce dense patterns of writes to all
74
Dell EMC Host Connectivity Guide for Windows
Virtual Provisioning
available storage, thin devices will tend to become fully allocated quickly. If thin
devices belonging to an over-subscribed pool are used in this type of environment,
out-of-space and undesired conditions may be encountered before an administrator
can take steps to add storage capacity to the thin data pool. Such environments are
called thin-hostile.
Thin-hostile environments
There are a variety of factors that can contribute to making a given application
environment thin-hostile, including:
◆
One step, or a combination of steps, involved in simply preparing storage for use by
the application may force all of the storage that is being presented to become fully
allocated.
◆
If the storage space management policies of the application and underlying
software components do not tend to reuse storage that was previously used and
released, the speed in which underlying thin devices become fully allocated will
increase.
◆
Whether any data copy operations (including disk balancing operations and
de-fragmentation operations) are carried out as part of the administration of the
environment.
◆
If there are administrative operations, such as bad block detection operations or file
system check commands, that perform dense patterns of writes on all reported
storage.
◆
If an over-subscribed thin device configuration is used with a thin-hostile
application environment, the likely result is that the capacity of the thin pool will
become exhausted before the storage administrator can add capacity unless
measures are taken at the host level to restrict the amount of capacity that is
actually placed in control of the application.
Pre-provisioning with thin devices in a thin hostile environment
In some cases, many of the benefits of pre-provisioning with thin devices can be
exploited in a thin-hostile environment. This requires that the host administrator
cooperate with the storage administrator by enforcing restrictions on how much
storage is placed under the control of the thin-hostile application.
For example:
◆
The storage administrator pre-provisions larger than initially needed thin devices to
the hosts, but only configures the thin pools with the storage needed initially. The
various steps required to create, map, and mask the devices and make the target
host operating systems recognize the devices are performed.
◆
The host administrator uses a host logical volume manager to carve out portions of
the devices into logical volumes to be used by the thin-hostile applications.
Implementation considerations
75
Virtual Provisioning
◆
The host administrator may want to fully preallocate the thin devices underlying
these logical volumes before handing them off to the thin-hostile application so
that any storage capacity shortfall will be discovered as quickly as possible, and
discovery is not made by way of a failed host write.
◆
When more storage needs to be made available to the application, the host
administrator extends the logical volumes out of the thin devices that have already
been presented. Many databases can absorb an additional disk partition
non-disruptively, as can most file systems and logical volume managers.
◆
Again, the host administrator may want to fully allocate the thin devices underlying
these volumes before assigning them to the thin-hostile application.
In this example it is still necessary for the storage administrator to closely monitor
the over-subscribed pools. This procedure will not work if the host administrators
do not observe restrictions on how much of the storage presented is actually
assigned to the application.
Host boot/root/swap/dump devices positioned on Symmetrix VP (tdev)
devices
A boot /root /swap /dump device positioned on Symmetrix Virtual Provisioning (thin)
device(s) is supported with Enginuity 5773 and later. However, some specific
processes involving boot /root/swap/dump devices positioned on thin devices should
not have exposure to encountering the out-of-space condition. Host-based processes
such as kernel rebuilds, swap, dump, save crash, and Volume Manager configuration
operations can all be affected by the thin provisioning out-of-space condition. This
exposure is not specific to Dell EMC's implementation of thin provisioning. Dell EMC
strongly recommends that the customer avoid encountering the out-of-space
condition involving boot / root /swap/dump devices positioned on Symmetrix VP (thin)
devices using the following recommendations:
◆
We strongly recommend that Virtual Provisioning devices utilized for boot
/root/dump/swap volumes be fully allocated or the VP devices must not be
oversubscribe.
Should you use an over-subscribed thin pool, you need to take the necessary
precautions to ensure that they do not encounter the out-of-space condition.
◆
We do not recommend implementing space reclamation, available with Enginuity
5874 and later, with pre-allocated or over-subscribed Symmetrix VP (thin) devices
that are utilized for host boot/root/swap/dump volumes. Although not
recommended, Space reclamation is supported on the listed types of volumes.
If you use space reclamation on this thin device, t be aware that this freed space
may ultimately be claimed by other thin devices in the same pool and may not be
available to that particular thin device in the future.
Cluster configurations
When using high availability in a cluster configuration, it is expected that no single
point of failure exists within the cluster configuration and that one single point of
failure will not result in data unavailability, data loss, or any significant application
becoming unavailable within the cluster. Virtual provisioning devices (thin devices) are
76
Dell EMC Host Connectivity Guide for Windows
Virtual Provisioning
supported with cluster configurations; however, over-subscription of virtual devices
may constitute a single point of failure if an out-of-space condition should be
encountered. To avoid potential single points of failure, appropriate steps should be
taken to avoid under-provisioned virtual devices implemented within high availability
cluster configurations.
Implementation considerations
77
Virtual Provisioning
Operating system characteristics
Most host applications will behave in a similar manner in comparison to the normal
devices when writing to thin devices. This same behavior can also be observed as long
as the thin device written capacity is less than thin device subscribed capacity.
However, issues can arise when the application writes beyond the provisioned
boundaries. With the current behavior of the Windows Operating System, the
exhaustion of the thin pool causes undesired results. Specifics are included below:
◆
Logical Volume Manager software SVM and VxVM
Cannot write to any volumes that are built on the exhausted pool.
◆
Windows NTFS File System
• The host reports the errors "File System is full" to the Windows system event
log. The larger the data file size that is being written to the thin device, the
more ‘file system is full’ error messages will be reported.
• The writing data file has corrupted data.
• Cannot create a file system on the exhausted pool.
• Cannot write a data file to the exhausted pool.
In the condition where the host is exposed to pre-provisioned thin devices that had not
been bound to the thin pool, the host may take a little longer time during boot up.
78
Dell EMC Host Connectivity Guide for Windows
CHAPTER 4
Invisible Body Tag
Windows Host
Connectivity
with Dell EMC
VPLEX
This chapter describes Dell EMC VPLEX. Topics include:
◆
◆
◆
◆
◆
◆
◆
Dell EMC VPLEX.................................................................................................. 80
Prerequisites ............................................................................... 81
Host connectivity....................................................................... 82
Configuring Fibre Channel HBAs ................................................ 83
Windows Failover Clustering with VPLEX ................................... 91
Setting up quorum on a Windows 2012/2012 R2 Failover
Cluster for VPLEX Metro or Geo clusters.................................. 125
Configuring quorum on Windows 2008/2008 R2 Failover
Cluster for VPLEX Metro or Geo clusters.................................. 131
Windows Host Connectivity with Dell EMC VPLEX
79
Windows Host Connectivity with Dell EMC VPLEX
Dell EMC VPLEX
For detailed information about VPLEX, refer to the following VPLEX documentation
available at Dell EMC Online Support for configuration and administration operations:
◆
EMC VPLEX with GeoSynchrony Product Guide
◆
EMC VPLEX with GeoSynchrony CLI Reference Guide
◆
EMC VPLEX with GeoSynchrony Security Configuration Guide
◆
EMC VPLEX Hardware Installation Guide
◆
EMC VPLEX Release Notes
◆
Implementation and Planning Best Practices for EMC VPLEX Technical Notes
◆
VPLEX online help, available on the Management Console GUI
◆
SolVe Desktop for VPLEX available at Dell EMC Online Support
◆
Dell EMC Simple Support Matrix, VPLEX and GeoSynchrony, available at Dell
EMC E-Lab Navigator
For the most up-to-date support information, refer to the Dell EMC Simple Support
Matrix.
80
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
Prerequisites
Before configuring VPLEX in the Windows environment, complete the following on
each host:
◆
◆
Confirm that all necessary remediation has been completed.
This ensures that OS-specific patches and software on all hosts in the VPLEX
environment are at supported levels according to the Dell EMC Simple Support
Matrix.
Confirm that each host is running VPLEX-supported failover software and has at
least one available path to each VPLEX fabric.
Note: Always refer to the Dell EMC Simple Support Matrix for the most up-to-date
support information and prerequisites.
◆
If a host is running PowerPath, confirm that the load-balancing and failover policy
is set to Adaptive.
Prerequisites
81
Windows Host Connectivity with Dell EMC VPLEX
Host connectivity
Consult the Implementation and Planning Best Practices for EMC VPLEX Technical
Notes available at Dell EMC Online Support for Windows host connectivity
recommendations and best practices with VPLEX configurations.
For the most up-to-date information on qualified switches, hosts, host bus adapters,
and software, refer to the always consult the Dell EMC Simple Support Matrix (ESM),
available through Dell EMC E-Lab Navigator (ELN), or contact your Dell EMC
Customer Representative.
The latest Dell EMC-approved HBA drivers and software are available for download at
the following websites:
◆
https://www.broadcom.com/
◆
http:/www.QLogic.com
◆
http://www.brocade.com
The Dell EMC HBA installation and configurations guides are available at the Dell
EMC-specific download pages of these websites.
Note: Direct connect from a host bus adapter to a VPLEX engine is not supported.
82
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
Configuring Fibre Channel HBAs
This section details Fibre Channel HBA-related configuration details that must be
addressed when using Fibre Channel with VPLEX.
IMPORTANT
The values provided are required and optimal for most scenarios. However, in host IO
profiles with large-block reads, the values may need to be tuned if the performance of
the VPLEX shows high front-end latency in the absence of high back-end latency,
which has visible impact on host application(s). This may be an indication that there
are too many outstanding IOs at a given time per port. If the recommended settings do
not perform well in the environment, contact Dell EMC Support for additional
recommendations.
For further information on how to monitor VPLEX performance, refer to the
"Performance and Monitoring" section of the VPLEX Administration Guide. If host
application(s) is seeing a performance issue with the required settings, contact Dell
EMC Support for further recommendations.
Setting queue depth and execution throttle for QLogic
Note: Changing the HBA queue depth is designed for advanced users. Increasing the
queue depth may cause hosts to over-stress other arrays connected to the Windows
host, resulting in performance degradation while performing IO.
The execution throttle setting controls the amount of outstanding I/O requests per
HBA port. The HBA execution throttle should be set to the QLogic default value, which
is 65535. This can be done at the HBA firmware level using the HBA BIOS or the
QConvergeConsole CLI or GUI.
The queue depth setting controls the amount of outstanding I/O requests per a single
path. On Windows, the HBA queue depth can be adjusted using the Windows Registry.
Note: When the execution throttle in the HBA level is set to a value lower than the
queue depth, it may limit the queue depth to a lower value than the set value.
The following procedures detail how to adjust the queue depth setting for QLogic
HBAs:
◆
“Setting the queue depth for the Qlogic FC HBA” on page 83
◆
“Setting the execution throttle on the Qlogic FC HBA” on page 84
◆
“Setting the queue depth and queue target on the Emulex FC HBA” on page 89
Follow the appropriate procedure according to the HBA type. For any additional
information, refer to the HBA vendor's documentation.
Setting the queue depth for the Qlogic FC HBA
1. On the desktop, click Start, select Run, and open the REGEDIT (Registry Editor).
Configuring Fibre Channel HBAs
83
Windows Host Connectivity with Dell EMC VPLEX
Note: Some driver versions do not create the registry by default. In these cases,
the user must manually create the registry.
2. Select HKEY_LOCAL_MACHINE and follow the tree structure down to the
QLogic driver as shown in the following figure and double-click DriverParameter:
The Edit String dialog box displays:
3. Change the value of qd to 20. The value is set in hexadecimal; 20 is 32 in decimal.
If additional driver parameters are already set, and the string qd= does not exist,
append the following text to the end of the string using a semicolon (";") to
separate the new queue depth value from previous entries:
;qd=20
4. Click OK.
The registry should appear as follows:
5. Exit the Registry Editor and reboot the Windows host.
Setting the execution throttle on the Qlogic FC HBA
1.
Install QConvergeConsole GUI or CLI.
2. Follow the GUI or CLI directions, outlined in this section to set the execution
throttle.
84
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
Using QConvergeConsole GUI
1. Start the QConvergeConsole GUI. The GUI displays, as shown in the following
figure:
2. Select one of the adapter ports in the navigation tree on the left.
3. Select Host > Parameters > Advanced HBA Parameters.
4. Set the Execution Throttle to 65535.
5. Click Save to save the settings.
6. Repeat the above steps for each port on each adapter connecting to VPLEX.
Using QConvergeConsole CLI
1. Select 2:Adapter Configuration from the main menu:
Configuring Fibre Channel HBAs
85
Windows Host Connectivity with Dell EMC VPLEX
2. Select 3. HBA Parameters:
3. Select the HBA (1: Port 1: in the following example):
4. Select 2. Configure HBA Parameters:
86
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
5. Select 11. Execution Throttle:
6. Set the value to 65535:
Note: The current value is in the second set of square brackets. The first is the
allowable range.
Configuring Fibre Channel HBAs
87
Windows Host Connectivity with Dell EMC VPLEX
7. Verify the options:
8. Validate that the Execution Throttle is set to the expected value of 65535:
9. Repeat the above steps for each port on each adapter connecting to VPLEX.
Setting queue depth and queue target for Emulex
Note: Changing the HBA queue depth is designed for advanced users. Increasing the
queue depth may cause hosts to over-stress other arrays connected to the Windows
host, resulting in performance degradation while performing IO.
Queue depth setting controls the amount of outstanding I/O requests per a single
LUN/target. On Windows, the Emulex HBA queue depth can be adjusted via the
Emulex UI (OneCommand).
Queue target controls I/O depth limiting on a per target or per LUN basis. If set to 0
= depth, limitation is applied to individual LUNs. If set to 1 = depth, limitation is applied
across the entire target. On Windows, the Emulex HBA queue depth can be adjusted
via the Emulex UI (OneCommand).
88
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
The following procedures detail adjusting the queue depth and queue target settings
for Emulex HBAs as follows:
◆
Set the Emulex HBA adapter queue depth in Windows to 32.
◆
Set the Emulex HBA adapter queue target in Windows to 0.
Note: This means 32 outstanding IOs per ITL, so if a host has 4 paths then there are 32
outstanding IOs per path, resulting in a total of 128.
Follow the appropriate procedure according to the HBA type. For any additional
information please refer to the HBA vendor's documentation.
Setting the queue depth and queue target on the Emulex FC HBA
Setting the queue depth on the Emulex FC HBA is done using Emulex UI
(OneCommand). OneCommand detects the active Emulex driver and enables changing
the relevant driver's parameters, specifically queue depth.
Note: Setting the queue-depth per this procedure is not disruptive.
To set the queue depth and queue target on the Emulex FC HBA:
1.
Install OneCommand.
2. Launch the OneCommand UI.
3. Select the relevant host name from the Hosts list.
4. Expand the HBA in the navigational tree and select the HBA port.
5. Select the Driver Parameters tab.
Configuring Fibre Channel HBAs
89
Windows Host Connectivity with Dell EMC VPLEX
6. In the Adapter Parameter list, locate the QueueDepth parameter and set its value
to 32.
7. In the same list, locate the QueueTarget parameter and set its value to 0.
8. Click Apply.
9. Repeat the above steps for each port on the host that has VPLEX Storage
exposed.
90
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
Windows Failover Clustering with VPLEX
Microsoft strongly recommends that before you create a failover cluster, you validate
your configuration; that is, run all tests in the Validate a Configuration Wizard. By
running these tests, you can confirm that your hardware and settings are compatible
with failover clustering.
IMPORTANT
With Windows Server 2012 or Windows Server 2012 R2, cluster validation storage
tests may not discover VPLEX distributed devices when servers are geographically
dispersed and configured on different VPLEX sites.
The reason for this is because a storage validation test selects only shared LUNs. A
LUN is determined to be shared if its disk signatures, device identification number
(page 0×83), and storage array serial number are the same on all cluster nodes.
When you have site-to-site mirroring configured (VPLEX distributed device), a LUN in
one site (site A) has a mirrored LUN in another site (site B). These LUNs have the same
disk signatures and device identification number (page 0×83), but the VPLEX storage
array serial numbers are different. Therefore, they are not recognized as shared LUNs.
The following is an example of what is reported in the cluster validation logs:
Cluster validation message: List Potential Cluster Disks
Description: List disks that will be validated for cluster compatibility. Clustered
disks which are online at any node will be excluded.
Start: 11/17/2013 5:59:01 PM.
• Physical disk 84d2b21a is visible from only one node and will not be tested.
Validation requires that the disk be visible from at least two nodes. The disk is
reported as visible at the following node:
WNH6-H5.elabqual.emc.com
• Physical disk 6f473a9f is visible from only one node and will not be tested.
Validation requires that the disk be visible from at least two nodes. The disk is
reported as visible at the following node:
WNH6-H13.elabqual.emc.com
To resolve the issue, run all the cluster validation tests before you configure distributed
devices to the geographically dispersed servers.
Note: If the validation test is needed later for support situations, LUNs that are not
selected for storage validation tests are supported by Microsoft and Dell EMC Shared
LUNs (distributed devices).
For more information, refer to the Microsoft KB article, Storage tests on a failover
cluster may not discover all shared LUNs.
Windows Failover Clustering with VPLEX
91
Windows Host Connectivity with Dell EMC VPLEX
Setting up quorum on a Windows 2012/2012 R2 Failover
Cluster for VPLEX Metro or Geo clusters
The recommended Windows Failover Clustering Quorum for stretched or cross
connected clusters is the File Share Witness. To set up the File Share Witness quorum,
complete the following steps.
1.
In Failover Cluster Manager, select the cluster and from the drop-down menu
select More Actions > Configure Cluster Quorum Settings, as follows:
The Configure Cluster Quorum Wizard displays:
2. Click Next.
92
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
The Select Quorum Configuration Option window displays:
3. Click the Select the quorum witness option and click Next.
The Select Quorum Witness window displays:
4. Select Configure a file share witness and click Next.
Setting up quorum on a Windows 2012/2012 R2 Failover Cluster for VPLEX Metro or Geo clusters
93
Windows Host Connectivity with Dell EMC VPLEX
The Configure a File Share Witness window displays:
For the server hosting the file share, follow these requirements and
recommendations:
• Must have a minimum of 5 MB of free space
• Must be dedicated to the single cluster and not used to store user or application
data
• Must have write permissions enabled for the computer object for the cluster
name
The following are additional considerations for a file server that hosts the file share
witness:
• A single file server can be configured with file share witnesses for multiple
clusters.
• The file server must be on a site that is separate from the cluster workload. This
enables equal opportunity for any cluster site to survive if site-to-site network
communication is lost. If the file server is on the same site, that site becomes
the primary site, and it is the only site that can reach the file share.
• The file server can run on a virtual machine if the virtual machine is not hosted
on the same cluster that uses the file share witness.
• For high availability, the file server can be configured on a separate failover
cluster.
5. Click Next.
94
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
The Confirmation screen displays:
6. Verify the settings and click Next.
The Summary window displays:
7. You can view this report or click Finish to complete the file share witness
configuration.
Setting up quorum on a Windows 2012/2012 R2 Failover Cluster for VPLEX Metro or Geo clusters
95
Windows Host Connectivity with Dell EMC VPLEX
Configuring quorum on Windows 2008/2008 R2 Failover
Cluster for VPLEX Metro or Geo clusters
This section contains the following information:
◆
“VPLEX Metro or Geo cluster configuration” on page 96
◆
“Prerequisites” on page 97
◆
“Setting up quorum on a Windows 2008/2008R2 Failover Cluster for VPLEX Metro
or Geo clusters” on page 97
VPLEX Metro or Geo cluster configuration
Two VPLEX Metro clusters, connected within metro (synchronous) distances of
approximately 60 miles (100 kilometers), form a Metro-Plex cluster. Figure 12 shows
an example of a VPLEX Metro cluster configuration. VPLEX Geo cluster configuration
is the same and adds the ability to dynamically move applications and data across
different compute and storage installations across even greater distances.
Figure 12 VPLEX Metro cluster configuration example
Note: All connections shown in Figure 12 are Fibre Channel, except the network
connections, as noted.
The environment in Figure 12 consists of the following:
◆
◆
96
Node-1 – Windows 2008 or Windows 2008 R2 Server connected to the VPLEX
instance over Fibre Channel.
Node -2 – Windows 2008 or Windows 2008 R2 Server connected to the VPLEX
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
◆
instance over Fibre Channel.
VPLEX instance – One or more engine VPLEX having a connection through the L2
switch to back-end and front-end devices.
Prerequisites
Ensure the following before configuring the VPLEX Metro or Geo cluster:
◆
VPLEX firmware is installed properly and the minimum configuration is created.
◆
All volumes to be used during the cluster test should have multiple back-end and
front-end paths.
Note: Refer to the Implementation and Planning Best Practices for EMC VPLEX
Technical Notes, available on Dell EMC Online Support, for best practices for the
number of paths for back-end and front-end paths.
◆
All hosts/servers/nodes of the same configuration, version, and service pack of the
operating system are installed.
◆
All nodes are part of the same domain and are able to communicate with each
other before installing Windows Failover Clustering.
◆
One free IP address is available for cluster IP in the network.
◆
PowerPath or MPIO is installed and enabled on all the cluster hosts.
◆
The hosts are registered to the appropriate View and visible to VPLEX.
◆
All volumes to be used during cluster test should be shared by all nodes and
accessible from all nodes.
◆
A network fileshare is required for cluster quorum.
Setting up quorum on a Windows 2008/2008R2 Failover Cluster for VPLEX
Metro or Geo clusters
To set up a quorum on VPLEX Metro or Geo clusters for Windows Failover Cluster,
complete the following steps.
1. Select the quorum settings. In the Failover Cluster Manager, right-click on the
cluster name and select More Actions > Configure Cluster Quorum Settings >
Node and File Share Majority:
Configuring quorum on Windows 2008/2008 R2 Failover Cluster for VPLEX Metro or Geo clusters
97
Windows Host Connectivity with Dell EMC VPLEX
The Node and File Share Majority model is recommended for VPLEX Metro and
Geo environments.
2. In the Configure Cluster Quorum Wizard, click Next:
98
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
3. In the Select Quorum Configuration window, ensure that the Node and File
Share Majority radio button is selected, and then click Next:
Configuring quorum on Windows 2008/2008 R2 Failover Cluster for VPLEX Metro or Geo clusters
99
Windows Host Connectivity with Dell EMC VPLEX
4. In the Configure File Share Witness window, ensure that the \\sharedfolder from
any Windows host in a domain other than the configured Windows Failover Cluster
nodes is in the Shared Folder Path, and then click Next:
5. In the Confirmation window, click Next to confirm the details:
100
Dell EMC Host Connectivity Guide for Windows
Windows Host Connectivity with Dell EMC VPLEX
6. In the Summary window, go to the Failover Cluster Manager and verify that the
quorum configuration is set to \\sharedfolder:
7. Click Finish.
Configuring quorum on Windows 2008/2008 R2 Failover Cluster for VPLEX Metro or Geo clusters
101
Windows Host Connectivity with Dell EMC VPLEX
102
Dell EMC Host Connectivity Guide for Windows
CHAPTER 5
Invisible Body Tag
Dell EMC
PowerPath for
Windows
This section provides information about PowerPath for Windows.
◆
◆
◆
PowerPath and PowerPath iSCSI....................................................................... 104
PowerPath for Windows .......................................................... 105
PowerPath verification and problem determination ................... 108
Dell EMC PowerPath for Windows
103
Dell EMC PowerPath for Windows
PowerPath and PowerPath iSCSI
PowerPath for Windows software is available in two different packages: PowerPath for
Windows and PowerPath iSCSI for Windows. It is important to know the differences
between the two packages before deploying the software:
PowerPath for Windows
PowerPath for Windows supports both Fibre Channel and iSCSI environments, and is
Microsoft digitally certified only for Fibre Channel environments. PowerPath for
Windows supports failover path management and load-balancing for up to 32 paths in
heterogeneous storage environments. PowerPath for Windows is not currently
supported by Microsoft for iSCSI implementations, although it is supported by Dell
EMC for Dell EMC iSCSI storage systems.
PowerPath iSCSI for Windows
PowerPath iSCSI for Windows supports EMC VNX series and CLARiiON iSCSI storage
systems, and is Microsoft digitally certified and is built on the Microsoft MPIO
framework. PowerPath iSCSI for Windows supports failover path management for up
to 8 paths in iSCSI storage environments.
104
Dell EMC Host Connectivity Guide for Windows
Dell EMC PowerPath for Windows
PowerPath for Windows
The following information is included in this section:
◆
“PowerPath and MSCS” on page 105
◆
“Integrating PowerPath into an existing MSCS cluster” on page 105
PowerPath and MSCS
If you are installing PowerPath and MSCS for the first time, install PowerPath first,
and then install MSCS. Installing PowerPath first avoids having to disrupt cluster
services at a later time.
Integrating PowerPath into an existing MSCS cluster
You can integrate PowerPath into an existing MSCS cluster without shutting down the
cluster, if there is close coordination between the nodes and the storage system. Each
node in a cluster can own a distinct set of resources. Node A is the primary node for its
resources and the failover node for Node B’s resources. Conversely, Node B is the
primary node for its resources and the failover node for Node A’s resources.
If after installing PowerPath on the cluster, you test node failover by disconnecting all
cables for a LUN or otherwise disrupting the path between the active host and the
array, Windows logs event messages indicating hardware or network failure and
possible data loss. If working correctly, the cluster will failover to a node with an active
path and you can ignore the messages from the original node as logged in the event
log. (You should check the application generating I/O to see if there are any failures. If
there are none, everything is working normally.)
Installing PowerPath in a clustered environment requires the following steps:
Moving resources to
Node A
◆
Move all resources to Node A
◆
Install PowerPath on Node B
◆
Configure additional paths between storage array and Node B
◆
Move all resources to Node B
◆
Install PowerPath on Node A
◆
Configure additional paths between storage array and Node A
◆
Return Node A’s resources back to Node A
To move all resources to Node A:
1. Start the MSCS Cluster Administrator utility, select Start, Programs,
Administrative Tools, Cluster Administrator.
2. In the left pane of the window, select all groups owned by Node B.
3. To move the resources to Node A, select File, Move Group. Alternatively, select
Move Group by right-clicking all group names under Groups in the left pane.
4. To pause Node b, click Node B and select File, Pause Node. This keeps the node
from participating in the cluster during PowerPath installation.
PowerPath for Windows
105
Dell EMC PowerPath for Windows
Installing PowerPath
onto Node B
To install PowerPath onto Node B:
1.
Install PowerPath.
2. Shut down Node B. In a cluster with greater than two nodes, install PowerPath on
these other nodes.
For example, in a four-node cluster, replace Node B with Nodes B, C, and D in step
4 of the previous section, “Moving resources to Node A,” and also in steps 1 and 2,
above.
Configuring additional
paths between the
storage system and
Node B
To configure additional paths:
1.
If necessary, reconfigure the storage system so its logical devices appear on
multiple ports.
2. If necessary, install additional HBAs on Node B.
3. Connect cables for new paths between Node B and the storage system.
4. Power on Node B.
5. To resume Node B, click Node B and select File, Resume Node.
In a cluster with greater than two nodes, configure additional paths between the
storage system and these other nodes. For example, in a four-node cluster, replace
Node B with Nodes B, C, and D in steps 2, 3, 4, and 5 above.
Moving resources to
Node B
To move all resources to Node B:
1.
In the left pane of the Cluster Administrator window, select all groups.
2. To move the resources to Node B, select File, Move Group.
In a cluster with greater than two nodes, move all resources to any of the remaining
nodes. For example, in a four-node cluster, replace Node B with any combination of
Nodes B, C, and D to which you want to move resources. For example, you could
move resources to Nodes B and C or move them to B, C, and D, or any permutation
of Nodes B, C, and D taken alone or together.
3. To pause Node A, click Node A and select File, Pause Node.
Installing PowerPath
onto Node A
To install PowerPath onto Node A:
1.
Install PowerPath.
2. Shut down Node A.
Configuring additional
paths between the
storage system and
Node A
To configure additional paths:
1.
If necessary, configure the storage system so its logical devices appear on multiple
ports.
2. If necessary, install additional HBAs on Node A.
3. Connect cables for new paths between Node A and the storage system.
4. Power on Node A.
5. To resume Node A, click Node A and select File, Resume Node.
106
Dell EMC Host Connectivity Guide for Windows
Dell EMC PowerPath for Windows
Returning Node A’s
resources to Node A
To return Node A’s resources:
1.
Using the MSCS Cluster Administrator utility, select all groups previously owned
by Node A.
2. To move the resources back to Node A, select File, Move Group.
PowerPath for Windows
107
Dell EMC PowerPath for Windows
PowerPath verification and problem determination
The following section assumes that PowerPath has been installed properly. Refer to
the appropriate PowerPath Installation and Administration Guide on Dell EMC Online
Support for instructions on how to install PowerPath. This section will help to verify
that PowerPath was installed correctly and help you to recognize some common
failure points.
Click the circled icon shown in Figure 13 to access the PowerPath Administration.
Figure 13 PowerPath Administration icon
Figure 14 shows the administration icon and the status it represents.
Figure 14 PowerPath Monitor Taskbar icons and status
108
Dell EMC Host Connectivity Guide for Windows
Dell EMC PowerPath for Windows
Figure 15 shows what PowerPath Administrator would look like if installed correctly.
Notice that in this case there is one path zoned between the HBA and one port on the
storage device.
Figure 15 One path
PowerPath verification and problem determination
109
Dell EMC PowerPath for Windows
When multiple paths are zoned to your storage device, PowerPath Administrator would
look like Figure 16:
Figure 16 Multiple paths
Problem determination
Determining the cause of loss if connectivity to the storage devices can be simplified
by using the PowerPath Administrator. Array ports that are offline, defective HBAs or
broken paths show up in the administrator GUI in various ways.
Table 1 on page 111 shows the known possible failure states. Referencing this table can
greatly reduce problem determination time.
110
Dell EMC Host Connectivity Guide for Windows
Dell EMC PowerPath for Windows
Table 1 Possible failure states
PowerPath verification and problem determination
111
Dell EMC PowerPath for Windows
Examples of some failures follow:
An error with an array port, or the path leading to the array port, is displayed in
Figure 17. This is symbolized by the red X through one of the array ports. Notice that
while the array port is down, access to the disk device is still available; degraded
access is noted by a red slash.
Figure 17 Error with an Array port
112
Dell EMC Host Connectivity Guide for Windows
Dell EMC PowerPath for Windows
Figure 18 shows the result of a problem with one of the HBA or the path leading to the
HBA. The failed HBA/path is marked with a red X. Again, notice that access to the disk
devices, while degraded, still exists.
Figure 18 Failed HBA path
Making changes to your environment
You must reconfigure PowerPath after making configuration changes that affect
host-to-storage-system connectivity or logical device identification, for example:
◆
Fibre Channel switch zone changes
◆
Adding or removing Fibre Channel switches
◆
Adding or removing HBAs or storage-system ports
◆
Adding or removing logical devices
In most cases making changes to your environment will be detected automatically by
PowerPath. Depending on the type HBA you are using you may have to scan for new
devices in device manager. On some occasions and depending on the operating system
version your are running you may need to reboot your system.
PowerPath messages
For a complete list of PowerPath messages and their meanings refer to the
"PowerPath Product Guide - PowerPath Messages" chapter for the version of
PowerPath you are running.
PowerPath verification and problem determination
113
Dell EMC PowerPath for Windows
114
Dell EMC Host Connectivity Guide for Windows
CHAPTER 6
Invisible Body Tag
Microsoft
Native MPIO
and Hyper-V
This section provides information for using Microsoft’s Native Multipath I/O (MPIO)
with Windows Server 2008, Windows Server 2008 R2, and Windows Server 2012.
◆
◆
◆
◆
Native MPIO with Windows Server 2008/Windows Server
2008 R2 ................................................................................... 100
Native MPIO with Windows Server 2012................................... 120
Known issues ............................................................................ 125
Hyper-V .................................................................................... 126
Microsoft Native MPIO and Hyper-V
115
Microsoft Native MPIO and Hyper-V
Native MPIO with Windows Server 2008/Windows Server
2008 R2
This section includes the following information:
◆
“Support for Native MPIO in Windows Server 2008 and Windows Server 2008 R2”
on page 116
◆
“Configuring Native MPIO for Windows 2008 Server Core and Windows 2008 R2
Server Core” on page 116
Support for Native MPIO in Windows Server 2008 and Windows Server 2008
R2
Windows Server 2008 and Windows Server 2008 R2 include native multipathing
(MPIO) support as a feature of the OS.
Native MPIO is supported with all Dell EMC storage arrays.
Note the following:
◆
For Windows Server 2008 Core and Windows 2008 R2 Server Core installations,
Native MPIO is failover only. There are no load balancing options available in the
default DSM for Dell EMC storage arrays.
◆
Default Microsoft MPIO Timer Counters are supported.
◆
Hosts running Windows Server 2008 and Windows Server 2008 R2 must be
manually configured so that the initiators are registered using failover mode 4
[ALUA].
◆
CLARiiON systems need to be on FLARE 26 or above to support Native MPIO.
◆
R30 is the minimum version supported with CX4.
◆
VNX OE for Block v31 is the minimum.
Configuring Native MPIO for Windows 2008 Server Core and Windows 2008
R2 Server Core
For Windows 2008 Server Core and Windows 2008 R2 Server Core, use the procedure
described in “Enabling Native MPIO on Windows Server 2008 Server Core and
Windows Server 2008 R2 Server Core” on page 119.
Note: Refer to Microsoft documentation for installing the Microsoft Multipath I/O
feature.
Native MPIO must be configured to manage VPLEX, Symmetrix DMX, VNX series, and
CLARiiON systems. Open Control Panel, then the MPIO applet.
The claiming of array/device families can be done in one of two ways as described in
“Method 1,” next, and in “Method 2” on page 117.
Method 1
116
Manually enter the Vendor and Device IDs of the arrays for native MPIO to claim and
manage.
Dell EMC Host Connectivity Guide for Windows
Microsoft Native MPIO and Hyper-V
Note: This may be the preferred method if all arrays are not initially connected during
configuration and subsequent reboots are to be avoided.
To manually enter the array vendor and product ID information:
1.
Use the MPIO-ed Devices tab in the MPIO Properties control panel applet.
2. Select Add and enter the vendor and product IDs of the array devices to be claimed
by native MPIO.
The vendor ID must be entered as a string of eight characters (padded with trailing
spaces) and followed by the product ID entered as a string of sixteen characters
(padded with trailing spaces).
For example, to claim a VNX series and CLARiiON RAID 1 device in MPIO, the string
would be entered as
DGC*****RAID*1**********
where the asterisk is representative of a space.
The vendor and product IDs vary based on the type of array and device presented to
the host, as shown in Table 2.
Table 2 Array and device types
Array type
LUN type
Vendor ID
Product ID
VPLEX VS1/VS2
Any
EMC
Invista
DMX, DMX-2, DMX-3, DMX-4, VMAX
40K, VMAX 20K/VMAX, VMAX 10K
(Systems with SN xxx987xxxx), VMAX
10K (Systems with SN xxx959xxxx), and
VMAXe
Any
EMC
SYMMETRIX
CX300, CX500, CX700, all CX3-based
array
AX4 Series, CX4 Series, CX3 Series,
VNX series and CLARiiON Virtual
Provisioning
JBOD (single disk)
DGC
DISK
RAID 0
DGC
RAID 0
RAID 1
DGC
RAID 1
RAID 3
DGC
RAID 3
RAID 5
DGC
RAID 5
RAID 6
DGC
VRAID
RAID 1/0
DGC
RAID 10
Method 2
Use the MPIO applet to discover, claim, and manage the arrays already connected
during configuration.
Note: This may be the preferred method if ease-of-use is required and subsequent
reboots are acceptable when each array is connected.
Native MPIO with Windows Server 2008/Windows Server 2008 R2
117
Microsoft Native MPIO and Hyper-V
IMPORTANT
MPIO limits the number of paths per LUN to 32. Exceeding this number will result in
the host crashing with a Blue Screen stop message. Do not exceed 32 paths per LUN
when configuring MPIO on your system.
Automatic discovery is configured using the Discover Multi-Paths tab of the MPIO
Properties control panel applet. Note that only arrays which are connected with at
least two logical paths will be listed as available to be added in this tab, as follows:
◆
Devices from VNX OE for Block v31 and CLARiiON systems (running FLARE R26 or
greater, configured for failover mode 4 [ALUA]) will be listed in the SPC-3
compliant section of the applet
◆
Devices from DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K (Systems with SN
xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and VMAXe arrays will be
listed in the Others section of the applet
◆
Devices from VPLEX arrays will be listed in the Others section of the applet
Select the array / device types to be claimed and managed by MPIO by selecting the
Device Hardware ID, and clicking the Add button.
Note: The OS will prompt you to reboot for each device type added. A single reboot
will suffice after multiple devices types are added.
Path management in Multipath I/O for VPLEX, Symmetrix DMX, VMAX 40K, VMAX 20K/VMAX,
VMAX 10K (Systems with SN xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and
VMAXe, VNX series, and CLARiiON systems
Following reboot, after all device types have been claimed by MPIO, each
VPLEX-based, Symmetrix DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K (systems
with SN xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and VMAXe-based,
VNX series-based, and CLARiiON-based disk will be shown in Device Manager as a
Multi-Path Disk Device. When managed by MPIO, a new tab, named MPIO, will be
available under Properties of the selected disk device. Under the MPIO tab, the
number of logical paths configured between the host and array should be reported.
The default Load Balance Policy (as reported in the MPIO tab) for each disk device
depends upon the type of disk device presented:
◆
118
In Windows server 2008, DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K
(systems with SN xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and
VMAXe devices will report a default Load Balance Policy as “Fail Over Only”,
where the first reported path is listed as “Active/Optimized” and all other paths
listed as “Standby.” In Windows server 2008 R2, DMX, VMAX 40K, VMAX
20K/VMAX, VMAX 10K (systems with SN xxx987xxxx), VMAX 10K (Systems with
SN xxx959xxxx), and VMAXe devices will report a default Load Balance Policy as
"Round Robin," where all the paths are listed as "Active/Optimized." The default
Dell EMC Host Connectivity Guide for Windows
Microsoft Native MPIO and Hyper-V
policy can be overridden by changing the Load Balance Policy to any available. See
the Windows Server 2008 and Windows Server 2008 R2 documentation for a
detailed description of available Load Balance Policies.
DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K (systems with SN xxx987xxxx),
VMAX 10K (Systems with SN xxx959xxxx), and VMAXe array devices attached to
the host with a default Load Balance Policy of “Fail Over Only”, can be overridden
by changing the Load Balance Policy to any available. See the Windows Server
2008 and Windows Server 2008 R2 documentation for a detailed description of
available Load Balance Policies. Note that the default Load Balance Policy
cannot be changed globally for all disk devices, the change must be done on a
per-disk device basis.
◆
VNX series and CLARiiON devices will report a default Load Balance Policy as
“Round Robin With Subset”, where all paths to the SP owning the device as
“Active/Optimized”, and all paths to the SP not owning the LUN as
“Active/Unoptimized”.
VNX series and CLARiiON devices attached to the host in ALUA mode (as is
required when using native MPIO) report the path state which is used directly by
the host running native MPIO and cannot be overridden by changing the Load
Balance Policy.
◆
VPLEX devices will report a default Load Balance Policy as "Round Robin" with
all active paths as "Active/Optimized". The default policy can be overridden by
changing the Load Balance Policy to any available, except "Fail Over Only". See
the Windows Server 2008 and Windows Server 2008 R2 documentation for a
detailed description of available Load Balance policies.
Note: The default Load Balance Policy cannot be changed globally for all disk
devices. The change must be done on a per-disk device basis.
Enabling Native MPIO on Windows Server 2008 Server Core and Windows Server 2008 R2
Server Core
MPIO and other features must be started from the command line since Windows
Server 2008 Server Core and Windows Server 2008 R2 Server Core are minimal
installations that do not have traditional GUI interfaces. Refer to
http://technet.microsoft.com for more information on Windows Server Core
installations.
To enable the native MPIO feature from the command line, type:
start /w ocsetup MultipathIo
After the system reboots, you can manage MPIO with the mpiocpl.exe utility. From
the command prompt, type:
mpiocpl.exe
The MPIO Properties window displays. From here, arrays/devices can be claimed and
managed as described in the section above for standard Windows installations.
For more information on Microsoft Native MPIO, refer to http://www.microsoft.com
and http://technet.microsoft.com.
Native MPIO with Windows Server 2008/Windows Server 2008 R2
119
Microsoft Native MPIO and Hyper-V
Native MPIO with Windows Server 2012
This section incudes the following information:
◆
“Support for Native MPIO in Windows Server 2012” on page 120
◆
“Configuring Native MPIO for Windows Server 2012” on page 120
Support for Native MPIO in Windows Server 2012
Windows Server 2012 include native multipathing (MPIO) support as a feature of the
OS.
Native MPIO is supported with Dell EMC CX4 Series, DMX-4, and VMAX storage array
models.
Note the following:
◆
To use the Microsoft default DSM, storage must be compliant with SCSI Primary
Commands-3 (SPC-3).
◆
Default Microsoft MPIO Timer Counters are supported.
◆
Hosts running Windows Server 2012 must be manually configured so that the
initiators are registered using failover mode 4 [ALUA].
◆
CLARiiON systems need to be on FLARE 30 or above to support Native MPIO.
◆
VNX OE for Block v31 is the minimum.
Configuring Native MPIO for Windows Server 2012
This section explains how to configure native MPIO for Dell EMC storage arrays.
Native MPIO is installed as an optional feature of the Windows Server 2012.
Note: Refer to Microsoft documentation for installing the Microsoft Multipath I/O
feature.
Configuring MPIO and installing DSM
When MPIO is installed, the Microsoft device-specific module (DSM) is also installed,
as well as an MPIO control panel. The control panel can be used to do the following:
◆
Configure MPIO functionality
◆
Install additional storage DSMs
◆
Create MPIO configuration reports
Opening the MPIO control panel
Open the MPIO control panel either by using the Windows Server 2012 control panel or
by using Administrative Tools.
To open the MPIO control panel using the Windows Server 2012 control panel,
complete the following steps:
1. On the Windows Server 2012 desktop, move your mouse to the lower left corner
and click Start.
120
Dell EMC Host Connectivity Guide for Windows
Microsoft Native MPIO and Hyper-V
2. Click MPIO.
To open the MPIO control panel using Administrative Tools, complete the following
steps
1. On the Windows Server 2012 desktop, move your mouse to the lower left corner
and click Start.
2. Point to Administrative Tools and click MPIO.
The MPIO control panel opens to the Properties dialog box.
Note: To access the MPIO control panel on Server Core installations, open a command
prompt and type MPIOCPL.EXE.
Once installed, native MPIO must be configured to manage VPLEX, Symmetrix DMX,
VNX series, and CLARiiON systems. Open Control Panel, then the MPIO applet.
Device discovery and claiming devices for MPIO
IMPORTANT
MPIO limits the number of paths per LUN to 32. Exceeding this number will result in
the host crashing with a Blue Screen stop message. Do not exceed 32 paths per LUN
when configuring MPIO on your system.
Automatic discovery is configured using the Discover Multi-Paths tab of the MPIO
Properties control panel applet. Note that only arrays which are connected with at
least two logical paths will be listed as available to be added in this tab, as follows:
◆
Devices from VNX OE for Block v31 and CLARiiON systems (running FLARE 30 or
greater, configured for failover mode 4 [ALUA]) will be listed in the SPC-3
compliant section of the applet.
◆
Devices from VNX OE for Block v31 and CLARiiON systems (running FLARE 30 or
greater, configured for failover mode 4 [ALUA]) will be listed in the SPC-3
compliant section of the applet.
◆
Devices from DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K (Systems with SN
xxx987xxxx), VMAX 10K (systems with SN xxx959xxxx), and VMAXe arrays will be
listed in the Others section of the applet.
◆
Devices from VPLEX arrays will be listed in the Others section of the applet.
Native MPIO with Windows Server 2012
121
Microsoft Native MPIO and Hyper-V
Select the array and device types to be claimed and managed by MPIO by selecting
the Device Hardware ID, and clicking Add.
Figure 19 MPIO Properties dialog box
Note: The OS will prompt you to reboot for each device type added. A single reboot
will suffice after multiple devices types are added.
Path management in Multipath I/O for VPLEX, Symmetrix DMX, VMAX 40K, VMAX 20K/VMAX,
VMAX 10K (Systems with SN xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and
VMAXe, VNX series, and CLARiiON systems
Following reboot, after all device types have been claimed by MPIO, each
VPLEX-based, Symmetrix DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K (systems
with SN xxx987xxxx), VMAX 10K (Systems with SN xxx959xxxx), and VMAXe-based,
VNX series-based, and CLARiiON-based disk will be shown in Device Manager as a
Multi-Path Disk Device.
When managed by MPIO, a new tab, MPIO, will be available under Properties of the
selected disk device. Under the MPIO tab, the number of logical paths configured
between the host and array should be reported.
This tab will also allow you to change the MPIO load balancing policy for a disk device.
Note: Some load balancing policies may not be available for specific array disk types.
For example, the Round Robin policy is not available for VNX disk devices, but Round
Robin with Subset is.
122
Dell EMC Host Connectivity Guide for Windows
Microsoft Native MPIO and Hyper-V
Options available for load balancing policies are as follows:
◆
Fail Over Only — Policy that does not perform load balancing. This policy uses a
single active path, and the rest of the paths are standby paths. The active path is
used for sending all I/O. If the active path fails, then one of the standby paths is
used. When the path that failed is reactivated or reconnected, the standby path
that was activated returns to standby.
◆
Round Robin — Load balancing policy that allows the Device Specific Module
(DSM) to use all available paths for MPIO in a balanced way. This is the default
policy that is chosen when the storage controller follows the active-active model
and the management application does not specifically choose a load-balancing
policy.
◆
Round Robin with Subset — Load balancing policy that allows the application to
specify a set of paths to be used in a round robin fashion, and with a set of standby
paths. The DSM uses paths from a primary pool of paths for processing requests as
long as at least one of the paths is available. The DSM uses a standby path only
when all the primary paths fail. For example, given 4 paths: A, B, C, and D, paths A,
B, and C are listed as primary paths and D is the standby path. The DSM chooses a
path from A, B, and C in round robin fashion as long as at least one of them is
available. If all three paths fail, the DSM uses D, the standby path. If paths A, B, or
C become available, the DSM stops using path D and switches to the available
paths among A, B, and C.
◆
Least Queue Depth — Load balancing policy that sends I/O down the path with
the fewest currently outstanding I/O requests. For example, consider that there is
one I/O that is sent to LUN 1 on Path 1, and the other I/O is sent to LUN 2 on Path
1. The cumulative outstanding I/O on Path 1 is 2, and on Path 2, it is 0. Therefore,
the next I/O for either LUN will process on Path 2.
◆
Weighted Paths —Load balancing policy that assigns a weight to each path. The
weight indicates the relative priority of a given path. The larger the number, the
lower ranked the priority. The DSM chooses the least-weighted path from among
the available paths.
◆
Least Blocks —Load balancing policy that sends I/O down the path with the least
number of data blocks currently being processed. For example, consider that there
are two I/Os: one is 10 bytes and the other is 20 bytes. Both are in process on Path
1, and both have completed Path 2. The cumulative outstanding amount of I/O on
Path 1 is 30 bytes. On Path 2, it is 0. Therefore, the next I/O will process on Path 2.
The default Load Balance Policy (as reported in the MPIO tab) for each disk device
depends upon the type of disk device presented:
Native MPIO with Windows Server 2012
123
Microsoft Native MPIO and Hyper-V
◆
In Windows Server 2012, DMX, VMAX 40K, VMAX 20K/VMAX, VMAX 10K
(systems with SN xxx987xxxx), VMAX 10K (systems with SN xxx959xxxx), and
VMAXe devices will report a default Load Balance Policy as "Round Robin,"
where all the paths are listed as "Active/Optimized."
◆
VNX series and CLARiiON devices will report a default Load Balance Policy as
“Round Robin With Subset”, where all paths to the SP owning the device as
“Active/Optimized”, and all paths to the SP not owning the LUN as
“Active/Unoptimized”.
VNX series and CLARiiON devices attached to the host in ALUA mode (as is
required when using native MPIO) report the path state, which is used directly by
the host running native MPIO and cannot be overridden by changing the Load
Balance Policy.
◆
VPLEX devices will report a default Load Balance Policy as "Round Robin" with
all active paths as "Active/Optimized."
Load balancing policies should be changed based on your particular environment. In
most cases, the default policy will be suitable for your I/O load needs. However, some
environments may require a change to the load balancing policy to improve
performance or better spread I/O load across storage front-end ports. Dell EMC does
not require a specific load balancing policy for any environment, and our customers are
free to make changes to their load balancing policies as they see fit to meet their
environment's needs.
For more information on Microsoft Native MPIO, refer to http://www.microsoft.com
and http://technet.microsoft.com.
124
Dell EMC Host Connectivity Guide for Windows
Microsoft Native MPIO and Hyper-V
Known issues
The following are known issues:
◆
When a Windows 2008 host with Native MPIO managing VNX series and CLARiiON
systems boots, MPIO will move all CLARiiON LUNs to a single Storage Processor
on the VNX series and CLARiiON system.
◆
Windows 2008 Native MPIO does not auto-restore a VNX series and CLARiiON
LUN to its default Storage Processor after any type of fault is repaired. For
example, after a non-disruptive upgrade of VNX series and CLARiiON software, all
VNX series and CLARiiON LUNs will be owned on a single Storage Processor.
• To address the above behavior, VNX series and CLARiiON management
software (Unisphere/Navisphere Manager or Navisphere Secure CLI) can be
used to manually restore LUNs to their default storage processor.
• Also, a VNX series and CLARiiON LUN can be assigned a Load Balance Policy of
Failover Only with the Preferred box selected on a path connected to the
default storage processor. Native MPIO will attempt to keep the preferred path
as active/optimized and will use that path for IO.
Only this single, preferred path will be used for IO; there is failover, but no
multipathing, under this Load Balance Policy. If the preferred path fails, Native
MPIO will select an alternate, healthy path for IO.
IMPORTANT
The implications of doing this should be clearly understood. There will be no
multipathing to this LUN if the above method is implemented – only failover.
Known issues
125
Microsoft Native MPIO and Hyper-V
Hyper-V
Hyper-V in Windows Server enables you to create a virtualized server computing
environment. This environment can improve the efficiency of your computing resources
by utilizing more of your hardware resources. This is made possible by using Hyper-V to
create and manage virtual machines and their resources. Each virtual machine is a
self-contained virtualized computer system that operates in an isolated execution
environment. This allows multiple operating systems to run simultaneously on one
physical computer. Hyper-V is an available role in Windows Server 2008 and later.
For information on Hyper-V, its many features and benefits, and installation
procedures, refer to http://www.microsoft.com and http://technet.microsoft.com.
126
Dell EMC Host Connectivity Guide for Windows
APPENDIX A
Invisible Body Tag
Persistent
Binding
This appendix provides information on persistent binding.
◆
Understanding persistent binding.............................................. 128
Persistent Binding
127
Persistent Binding
Understanding persistent binding
Persistent binding is the mechanism to create a continuous logical route from a storage
device object in the Windows host to a volume in the EMC storage array across the
fabric.
Without a persistent binding mechanism, the host cannot maintain persistent logical
routing of the communication from a storage device object across the fabric to an Dell
EMC storage array volume. If the physical configuration of the switch is changed (for
example, the cable is swapped or the host is rebooted), the logical route becomes
inconsistent, causing possible data corruption if the user application is modifying data
through inconsistent logical routing of the communication from the driver entry point
to a volume in an Dell EMC storage array across the fabric.
The Windows OS does not provide a satisfactory means to allow persistent binding.
Most software applications access storage using file systems managed by the
Windows OS. (File systems are represented as <drive letter><colon>, that is, C:, D:,
and so forth) For storage devices containing file systems, Windows writes a disk
signature to the disk device. The operating system can then identify, and associate
with, a particular drive letter and file system.
Since the signature resides on the disk device, changes can occur on the storage end
(a cable swap, for example) that can cause a disk device to be visible to the host server
in a new location. However, the OS looks for the disk signature and, providing that
nothing on the disk changed, associate the signature with the correct drive letter and
file system. This mechanism is strictly an operating system feature and is not
influenced by the Fibre Channel device driver.
Some software applications, however, do not use the Windows file systems or drive
letters for their storage requirements. Instead they access storage drives directly, using
their own built-in “file systems.” Devices accessed in this way are referred to as raw
devices and are known as physical drives in Windows terminology.
The naming convention for physical drives is simple and is always the same for
software applications using them. A raw device under Windows is accessed by the
name \\.\PHYSICALDRIVEXXX, where XXX is the drive number.
For example, a system with three hard disks attached using an Emulex Fibre Channel
controller assigns the disks the names \\.\PHYSICALDRIVE0, \\.\PHYSICALDRIVE1,
and \\.\PHYSICALDRIVE2. The number is assigned during the disk discovery part of
the Windows boot process.
During boot-up, the Windows OS loads the driver for the storage HBAs. Once loaded,
the OS performs a SCSI Inquiry command to obtain information about all the attached
storage devices. Each disk drive that it discovers is assigned a number in a semi-biased
first come, first serve fashion based on HBA. Semi-biased means the Windows system
always begins with the controller in the lowest-numbered PCI slot where a storage
controller resides. Once the driver for the storage controller is loaded, the OS selects
the adapter in the lowest-numbered PCI slot to begin the drive discovery process.
It is this naming convention and the process by which drives are discovered that makes
persistent binding (by definition) impossible for Windows. Persistent binding requires a
continuous logical route from a storage device object in the Windows host to a volume
in an Dell EMC storage array across the fabric. As mentioned above, each disk drive is
assigned a number in a first-come, first-serve basis. This is where faults can occur.
128
Dell EMC Host Connectivity Guide for Windows
Persistent Binding
Example
Imagine this scenario: A host system contains controllers in slots 0, 1, and 2. Someone
removes a cable from the Emulex controller in host PCI slot 0, then reboots the host.
During reboot, the Windows OS loads the Emulex driver during reboot and begins disk
discovery. Under the scenario presented above, there are no devices discovered on
controller 0, so the OS moves to the controller in slot 1 and begins naming the disks it
finds, starting with \\.\PHYSICALDRIVE0. Any software applications accessing
\\.\PHSYICALDRIVE0 before the reboot will be unable to locate their data on the
device, because it changed.
Figure 20 on page 129 shows the original configuration before the reboot. HBA0 is in
PCI slot 0 of the Windows host. Each HBA has four disk devices connected to it, so
Windows has assigned the name \\.\PHYSICALDRIVE0 to the first disk on HBA0.
Each disk after that is assigned a number in sequence as shown in Figure 20.
PHYSICALDRIVE0
HBA 0
Windows
Host
PHYSICALDRIVE4
HBA 1
HBA 2
PHYSICALDRIVE8
Figure 20 Original configuration before the reboot
Figure 21 shows the same host after the cable attached to HBA0 has been removed
and the host rebooted. Since Windows was unable to do a discovery on HBA0, it
assigned \\.\PHYSICALDRIVE0 to the first device it discovered. In this case, that first
device is connected to HBA1. Due to the shift, any software application accessing
\\.\PHYSICALDRIVE0 will not find data previously written on the original
\\.\PHYSICALDRIVE0.
HBA 0
Windows
Host
PHYSICALDRIVE0
HBA 1
HBA 2
PHYSICALDRIVE4
Figure 21 Host after the reboot
Note: Tape devices are treated the same as disk devices in Windows with respect to
persistent binding. Refer to your tape device documentation for more information.
Understanding persistent binding
129
Persistent Binding
Methods of persistent binding
The Windows device naming convention and disk discovery process does not allow the
Windows operating system to establish persistent binding.
◆
Dell EMC Volume Logix — Provides persistent binding through centralized
control by the Symmetrix Fibre Channel fabric ports.
For more information, refer to the EMC Volume Logix Product Guide.
◆
Switch zone mapping — Provides persistent binding through centralized control
by the Fibre Channel switch.
For more information, refer to the EMC Connectrix Enterprise Storage Network
System Planning Guide.
◆
(Emulex HBAs only) Emulex configuration tool mapping — Provides persistent
binding of targets through centralized control by the Emulex host adapter. This
requires the user to modify the mapping manually.
For more information, refer to EMC Host Connectivity with Emulex Fibre Channel
Host Bus Adapters (HBAs) and Converged Network Adapters (CNAs) in the
Windows Environment, available in the Dell EMC OEM section of the Broadcom
website or on Dell EMC Online Support.
a. Click drivers, software and manuals at the left side of the screen.
b. Click EMC at the upper center of the next screen.
c. Click the link to your HBA at the left side of the screen.
d. Under EMC Drivers, Software and Manuals, click the Installation and
Configuration link under Drivers for Windows <version>.
130
Dell EMC Host Connectivity Guide for Windows
APPENDIX B
Invisible Body Tag
Dell EMC
Solutions
Enabler
This appendix describes Dell EMC Solutions Enabler and migration considerations.
◆
Dell EMC Solutions Enabler....................................................... 132
Dell EMC Solutions Enabler
131
Dell EMC Solutions Enabler
Dell EMC Solutions Enabler
The Solutions Enabler SYMCLI is a specialized library consisting of commands that can
be invoked via the command line or within scripts. These commands can be used to:
◆
Monitor device configuration and status
◆
Perform control operations on devices and data objects within your managed
storage complex.
The target storage environments are typically Symmetrix-based, through some
features are supported for Unity and VNX systems as well.
For more information, refer to the Solutions Enabler Array Controls and Management
8.3.0 CLI User Guide, available on Dell EMC Online Support.
References
More information can be found in the following guides, available on Dell EMC Online
Support:
◆
For HBA configurations, see the appropriate host bus adapter guides:
• EMC Host Connectivity with Emulex Fibre Channel Host Bus Adapters (HBAs)
and Converged Network Adapters (CNAs) in the Windows Environment,
available on
• EMC Host Connectivity with QLogic Fibre Channel and iSCSI Host Bus
Adapters (HBAs) and Converged Network Adapters (CNAs) in the Windows
Environment
• EMC Host Connectivity with Brocade Fibre Channel and Host Bus Adapters
(HBAs) and Converged Network Adapters (CNAs) in the Windows Environment
• For additional migration considerations, refer to EMC's Data Migration All
Inclusive guide.
132
Dell EMC Host Connectivity Guide for Windows
APPENDIX C
Invisible Body Tag
Veritas Volume
Management
Software
Veritas Storage Foundation/Volume Manager from Symantec replaces the native
volume manager of the Windows operating system to allow management of physical
disks as logical devices. The added features of this software are designed to enhance
data storage management by controlling space allocation, performance, data
availability, device installation, and system monitoring of private and shared systems.
Refer to http://www.symantec.com for more information about Veritas Storage
Foundation, documentation, and software availability.
Refer to the latest Dell EMC Simple Support Matrix to determine which Veritas
Storage Foundation/Volume Manager configurations are supported and what service
packs may be required for your configuration.
Veritas Volume Management Software
133