Professional Documents
Culture Documents
v4.0
EMC Corporation
Corporate Headquarters:
Hopkinton, MA 01748-9103
1-508-435-1000
www.EMC.com
Contents
Preface.................................................................................................................................................... 13
Chapter 1
Introduction
VPLEX overview.....................................................................................................
Major components ............................................................................................
Preparing for a new installation ...........................................................................
Required information.......................................................................................
Required hardware...........................................................................................
Required software ............................................................................................
Required tools ...................................................................................................
Reference: VPLEX requirements ....................................................................
Chapter 2
16
18
20
20
20
21
21
21
Chapter 3
Setting up VPLEX
Preliminary tasks .................................................................................................... 34
Setting up a VPLEX Local implementation ........................................................ 35
Task 1: Set the management servers public IP address ............................. 35
Task 2: Launch the VPLEX EZ-Setup Wizard .............................................. 42
Task 3: Verify the product version ................................................................. 42
Task 4: Verify the VPLEX directors ................................................................ 42
Task 5: Open a second PuTTY session ........................................................... 43
Task 6: Verify the director count .................................................................... 43
Task 7: Verify the Fibre Channel switches if present .................................. 43
Task 8: Expose back-end storage .................................................................... 43
Task 9: Verify storage volume availability ................................................... 44
Task 10: Create a meta-volume ....................................................................... 44
Task 11: Determine the TLA ........................................................................... 45
Task 12: Configure Call Home ........................................................................ 46
Appendix A
Reference
IP addresses and component IDs ......................................................................... 98
Internal cabling ..................................................................................................... 100
Quad-engine cluster ....................................................................................... 101
Dual-engine cluster ........................................................................................ 105
Single-engine cluster ...................................................................................... 109
Cabinet power connections ................................................................................. 111
Figures
Title
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Page
16
17
18
19
22
24
24
25
26
27
28
28
29
30
30
31
31
31
32
32
35
36
37
37
38
39
40
41
41
51
55
56
57
58
59
60
62
62
63
63
Figures
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Tables
Title
1
2
Page
Tables
10
Attention
Warnung
Nur fr Fachpersonal.
Dieses EMC Produkt verfgt ber mehrere elektrische Netzanschlsse. Zur
Vermeidung eines elektrischen Schlages sind vor Servicearbeiten an der
Stromversorgung alle Netzanschlsse zu trennen.
Kontinuierliche Erdung ist notwendig whrend der gesamten Betriebsdauer des
Gertes. Es ist unzulssig das Gert ohne Erdung zu betreiben. Gert muss geerdet
werden, bevor es am Stromnetz angeschlossen wird.
Additional warnings
and cautions
Before attempting to service the EMC hardware described in this document, observe
the following additional warnings and cautions:
WARNING
The hardware enclosure contains no user-serviceable parts, so it should not be moved
or opened for any reason by untrained persons. If the hardware needs to be relocated
or repaired, only qualified personnel familiar with safety procedures for electrical
equipment and EMC hardware should access components inside the unit or move the
unit.
11
WARNING
This product operates at high voltages. To protect against physical harm, power off
the system whenever possible while servicing.
WARNING
In case of fire or other emergency, isolate the system's power involved and alert
appropriate personnel.
CAUTION
Trained personnel are advised to exercise great care at all times when working on
the EMC hardware.
Remember to:
Static precautions
Remove rings, watches, or other jewelry and neckties before you begin any
procedures.
Use caution near any moving part and any part that may start unexpectedly
such as fans, motors, solenoids, etc.
EMC incorporates state-of-the-art technology in its designs, including the use of LSI
and VLSI components. These chips are very susceptible to damage caused by static
discharge and need to be handled accordingly.
CAUTION
Before handling printed circuit boards or other parts containing LSI and/or VLSI
components, observe the following precautions:
12
Preface
As part of an effort to improve and enhance the performance and capabilities of its product
lines, EMC periodically releases revisions of its hardware and software. Therefore, some
functions described in this document may not be supported by all versions 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,
please contact your EMC representative.
Audience
This document is intended for use by VPLEX customers and/or EMC field
personnel as an aid in setting up a new VPLEX installation.
Persons performing the procedures described in this document should be familiar
with Linux commands and text editors.
Related
documentation
Conventions used in
this document
Implementation and Planning Best Practices for EMC VPLEX Technical Notes
Customer Service Procedure: EMC Secure Remote Support Gateway, available in the
Application Download Area on the Global Services website
13
Preface
CAUTION
A caution contains information essential to avoid data loss or damage to the system
or equipment.
IMPORTANT
An important notice contains information essential to operation of the software.
WARNING
A warning contains information essential to avoid a hazard that can cause severe
personal injury, death, or substantial property damage if you ignore the warning.
Typographical conventions
EMC uses the following type style conventions in this document:
Bold
Italic
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
Courier italic
<>
[]
Your suggestions will help us continue to improve the accuracy, organization, and
overall quality of the user publications. Please send your opinion of this document to:
techpubcomments@EMC.com
14
1
Introduction
Introduction
15
Introduction
VPLEX overview
Figure 1 through Figure 3 show the main hardware components in the three available
configurations of an EMC VPLEX cluster (a single-cabinet configuration). A
VPLEX implementation can be VPLEX Local (single cluster) or VPLEX Metro
(two clusters, separated by up to 5 ms round trip time, or RTT).
ON
I
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
Management server
ON
I
O
OFF
Engine 1
Director 1B
Director 1A
SPS 1
Zep-008
Figure 1
16
Introduction
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
O
OFF
ON
I
ON
I
Management server
Engine 2
Director 2B
Director 2A
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
SPS 2
Engine 1
Director 1B
Director 1A
SPS 1
Zep-07
Figure 2
VPLEX overview
17
Introduction
Director 4B
Director 4A
Engine 4
ON
I
ON
I
O
OFF
O
OFF
SPS 4
Engine 3
SPS 3
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
Director 3B
Director 3A
COM switch B
UPS B
COM switch A
UPS A
O
OFF
O
OFF
ON
I
ON
I
Management server
Engine 2
Director 2B
Director 2A
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
SPS 2
Engine 1
Director 1B
Director 1A
SPS 1
Zep-06
Figure 3
Major components
Table 1
18
Component
Description
Engine (shown in
Figure 4 on page 19)
Standby power
supply (SPS)
Provides backup power (through a pair or redundant modules) that allows an engine to ride through
transient (30 seconds or less) power loss.
Introduction
Table 1
Component
Description
Provide highly available connectivity among the directors in a medium (two-engine) or large
(four-engine) cluster.
Uninterrupted power
supplies (UPS)
Provide backup power to the COM switches and management server in a medium or large cluster,
allowing the connected components to continue operating through a power loss of up to 30 seconds.
Management server
Provides management to the other components in the cluster, and provides an interface for access
from a remote management station.
Management server functions include:
Hosting VPLEX software.
Providing an interface between the customer and the VPLEX cluster.
Generating Call Home events.
Enabling secure management between clusters in a Metro-Plex.
Protecting the internal management network from the customer LAN.
Providing access to VPLEX components for field-replaceable unit (FRU) replacement procedures.
Facilitating nondisruptive upgrades (NDU)
Figure 4
Engine components
VPLEX overview
19
Introduction
Verify that all necessary hardware has been received at the customer site(s), as
documented in the purchase order and shipping list.
Verify that two storage devices of at least 78 GB each have been allocated for use
as a VPLEX meta-volume and a mirrored copy. (The meta-volume will contain
RAID and virtual volume metadata.)
Each storage device should be from a different array.
Note: If the meta-volume is on an EMC CLARiiON array, there must be active paths to
all VPLEX directors from the meta-volume LUN.
You must ensure that these storage devices will provide a maximum average
response time of 10 ms (for random 4K reads). The best practice to ensure this is
to allocate dedicated spindles on the array for any LUN that will be used as part
of the meta-volume mirror.
Verify that you have the following EMC Secure Remote Support (ESRS) gateway
information:
ESRS gateway account.
Model Number VS1
Serial Number Top-level assembly number (TLA) of the cluster, available
on the serial number tag attached to the top rear of the cabinet frame member
(and on the packing slip)
Site ID Provided by ESRS
IP address Address of the public Ethernet port on the VPLEX management
server
Required
information
Confirm that you have completed the VPLEX Pre-installation Configuration Worksheet,
available with other VPLEX documentation on EMC Powerlink.
Required hardware
You must have a laptop, desktop, PC, or similar device with a keyboard and
graphical interface for connecting to the VPLEX management server and configuring
your VPLEX cluster(s). For convenience, this guide uses the term laptop to describe
the device.
Before starting the installation and configuration procedures described in this guide,
confirm that your laptop has the following:
20
USB port.
CD-ROM drive.
Introduction
Required software
Fully charged batteries For mobility between the lab environment and a
quieter location from which you can perform most of the configuration tasks.
Generally, access to power outlets will not be a problem. Just in case, however,
fully charged batteries could be useful.
Windows OS
Required tools
Reference: VPLEX
requirements
You will need the following tools when unpacking and installing a VPLEX cluster:
8 mm Allen wrench or cabinet door key (to open and secure the doors)
Ensure that each host in a storage view has paths to at least two directors in each
cluster.
Ensure that multi-pathing is configured in such a way that the paths are
distributed evenly between directors A and B in each engine.
21
Introduction
Figure 5
22
2
Installing the
Hardware
This chapter describes the tasks to install the VPLEX hardware at a customer site.
Note: Although much of this guide is intended for use by customers, this chapter is for use
primarily by EMC personnel.
23
Installing a cluster
Task 1: Unpack and
set up the cabinet
Follow the steps in the EMC 40U-C Cabinet Unpacking and Setup Guide, which
accompanies each VPLEX cabinet (and is available on Powerlink).
Director n B
Director n A
I/O module carrier
Figure 6
Zep-041
If any of these components is not fully seated, loosen the orange captive screws,
push the component firmly into the backplane, and retighten the thumbscrews.
2. Confirm that all I/O modules are fully seated.
3. Confirm that all thumbscrews are finger-tight.
4. Confirm that all cables in the cabinet (except AC cables to site power) are securely
connected, and that the wire bail on each SPS power cable, as well as (if present)
each UPS and Fibre Channel switch power cable, is secure.
The bail on each SPS and UPS cable must be clamped onto the plugs plastic
molding, and the bail on each Fibre Channel switch must be clamped onto the
cable itself, as shown in Figure 7.
SYM-002500z
Figure 7
24
5. From the front of the cabinet, remove each engines bezel as shown inFigure 8 ,
and verify that the two power supply modules, four fan modules, and solid state
drive (SSD) enclosure are fully inserted and secure.
Power supply
Fan
SSD enclosure
CL4341c
Figure 8
Transit/storage environment
Condition
Temperature range
Relative humidity
Hours required
before applying
power
Nominal
30% or greater
Cold / Dry
Hot / Dry
Hot / Humid
30% to 45%
45% to 60%
60% to 80%
16
Note: Temperature and humidity values must be such that condensation does not occur on any
VPLEX component. Altitude and atmospheric pressure specifications are referenced to a
standard day at 58.7F (14.8C). Maximum wet bulb temperature is 82F (28C).
Installing a cluster
25
Figure 9
Connect customer-supplied Fibre Channel cables from the front-end and back-end
SANs to the appropriate VPLEX I/O ports (identified in Figure 9).
Route the cables horizontally to the sides of the cabinet as shown in Figure 10 on
page 27, to prevent blocking other components in the cabinet. Route the cables down
the sides of the cabinet through the plastic cable holders, and secure the cables with
the Velcro straps that are attached to the cabinet frame. Maintain a minimum radius
of two inches for any bend in a Fibre Channel cable.
26
Front-end
cables
Back-end
cables
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
Engine 4
Engine 3
ON
I
ON
I
O
OFF
O
OFF
Engine 2
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
Engine 1
Zep-005_FC
Figure 10
Installing a cluster
27
Task 5: Connect to
the customer
network
Connect the management server to the customer IP network as shown in Figure 11.
Customer
IP network
zep-09a
Figure 11
Task 6: Metro-Plex
only: Cable the
clusters
In a Metro-Plex configuration, connect each director in each cluster (through its WAN
ports on the I/O module carrier) to the intercluster SAN as shown in Figure 12:
1. Label each WAN cable at both ends to identify the connections.
2. Secure the WAN cables in the cable holders below the engines as shown in
Figure 13 on page 29, and route the cables up or down the sides of the cabinet
with the I/O cables.
B4-FC02
B4-FC03
A4-FC02
Intercluster
COM SAN
switch 1B
Figure 12
28
B4-FC02
A4-FC03
Intercluster
COM SAN
switch 1A
Intercluster cabling
ISL 1
ISL 2
Intercluster
COM SAN
switch 2A
Intercluster
COM SAN
switch 2B
B4-FC03
A4-FC02
A4-FC03
Zep-049
Separate wires
around posts
0
1
2
3
3
Detail
0
1
0
0
1
1
2
2
3
3
0
2
e
ul
od
M
A
0
0
I/O
1
1
2
2
3
3
zep-070a
1. Verify that all PDP switches and PDU circuit breakers (shown in Figure 14 on
page 30) are in the OFF position. As the figure notes, the upper pair of PDUs are
installed upside-down from the lower pair, and the upper PDPs are not used.
2. Complete the EMC AC Power Configuration Worksheet, and place it in a convenient
location foe use in verifying the data center power schemes.
3. Connect the main AC power cables from the lower PDPs into the sites power
sources. To ensure high availability, each cable must connect to a different power
source, as described in the EMC Best Practices Guide for AC Power Connections in
Two-PDP Bays.
Installing a cluster
29
Notes:
The upper PDPs are installed,
but are not used.
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
PDU circuit
breakers
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
PDP power
switch
To 30 A, 220 VAC
power source 2
Figure 14
PDP power
switch
To 30 A, 220 VAC
power source 1
Zep-013
CL4171
Figure 15
30
Generally, the SPS On-Battery LED stays on while the SPS units fully charge
(which could be a few minutes or a few hours, depending on the state of the
battery). If any amber LED not related to the SPS recharge remains on for
more than 10 minutes, verify that the components are cabled correctly.
Director LEDs
Steady green
Amber during POST, off
during normal operation
SPS LEDs:
On-line Enabled (LED on) or On-line Charging (LED flashing)
On-Battery
Replace Battery
Internal Check
Figure 16
Zep-011a
Overload LED
On battery LED
Power button
Figure 17
sym-001521z
b. Verify that neither circuit breaker (shown in Figure 18) on either UPS has
triggered. If a circuit breaker has triggered, press it to reseat it.
Circuit breakers
Zep-036a
Figure 18
Installing a cluster
31
c. Verify that the Power and Status LEDs (shown in Figure 19) on both Fibre
Channel switches are steady green.
Status LED
Power LED
Figure 19
Zep-034
Power button
Figure 20
Power LED
32
Zep-015
3
Setting up VPLEX
This chapter describes the tasks to set up a VPLEX Local or VPLEX Metro
implementation. Topics include:
Preliminary tasks........................................................................................................... 34
Setting up a VPLEX Local implementation ............................................................... 35
Setting up a VPLEX Metro implementation.............................................................. 55
Setting up VPLEX
33
Setting up VPLEX
Preliminary tasks
Before you begin the VPLEX setup tasks, perform the following steps:
Note: The values (especially version numbers) in any command output in this document that is
identified as an example might not match what you see on a customer system.
1. Confirm that you know where to find the following documentation on EMC
Powerlink:
Best Practices white papers.
EMC VPLEX Pre-installation Configuration Worksheet. This document is in
Microsoft Word format, to provide a means of typing information (such as IP
addresses) that you will need during the setup procedure.
EMC VPLEX CLI Guide, which provides detailed information on the VPlexcli
commands identified in this installation and setup guide.
VPLEX Procedure Generator, which contains procedures to upgrade to a new
GeoSynchrony (VPLEX operating system) release, and to troubleshoot
installation or configuration issues.
2. If not already done, download the EMC VPLEX Pre-installation Configuration
Worksheet and add the required information.
3. Install and launch the VPLEX Procedure Generator as directed on Powerlink.
Then:
a. Generate a troubleshooting document, for reference if necessary during the
setup procedure. In the procedure generator, follow the link VPLEX >
Troubleshooting > Installation and setup issues.
b. If you are going to upgrade GeoSynchrony to a later release, generate an
upgrade document. In the procedure generator, follow the link VPLEX >
Upgrade Procedures > Upgrade to VPLEX release <number>.
CAUTION
If you want to upgrade to a later GeoSynchrony release, you must perform the
entire 4.0 setup procedure as documented in this guide, and then perform an
upgrade to the desired release.
Note: If any operation described in this document fails, review the troubleshooting
document. If this does not resolve the problem, contact the EMC Support Center.
Next step
34
Setting up VPLEX
CAUTION
If you are setting up a VPLEX Metro implementation, follow the tasks in Setting
up a VPLEX Metro implementation on page 55. You cannot set up each cluster
individually and then join them later.
Before you can log in to the VPLEX management server over the customer network,
you must set the address on the management servers public Ethernet port:
Note: The tasks in this document use PuTTY for logging in to the management server.
However, you can use any similar Telnet/SSH client.
1. Connect your laptop to the management server as shown in Figure 21, using the
red service cable that hangs inside the EMC cabinet.
Laptop
Service port
Mgmt-B port
Figure 21
Mgmt-A port
zep-009b
2. On your laptops toolbar, select Start > Settings > Control Panel > Network
Connections (double-click).
3. If any public network connection is enabled on your laptop, right-click the
network card and select Disable.
4. On the Network Connections window, right-click Local Area Connection and
select Properties.
5. On the Local Area Connection Properties window, double-click Internet
Protocol (TCP/IP) under This connection uses the following items.
6. On the Internet Protocol (TCP/IP) Properties window:
a. If your laptop is configured with specific addresses, make a note of them, to
ensure that you reset the configuration correctly later.
b. Select Use the following IP addresses, and type the following values:
IP address:
128.221.252.3
Subnet mask:
255.255.255.224
Default gateway: 128.221.252.2
7. Click OK on the Internet Protocol (TCP/IP) Properties window, and again on the
Local Area Connection Properties window.
8. Close the Network Connections window.
35
Setting up VPLEX
9. Open a Command Prompt window (Start > Programs > Accessories > Command
Prompt), and type the following command to confirm the settings:
ipconfig
Output example:
Ethernet adapter Local Area Connection:
Connection-specific
IP Address. . . . .
Subnet Mask . . . .
Default Gateway . .
DNS
. .
. .
. .
Suffix
. . . .
. . . .
. . . .
.
.
.
.
:
:
:
:
eng.emc.com
128.221.252.3
255.255.255.224
128.221.252.2
10. Type the following command to verify that you can reach the management
servers service port:
ping 128.221.252.2
Type:
128.221.252.2
(default)
Type session name:
VPLEX_service
(default)
PuTTY
Figure 22
13. Select SSH in the Category list, and ensure that Preferred SSH protocol version
is set to 2, as shown in Figure 23.
36
Setting up VPLEX
SSH protocol
version
putty_ssh
Figure 23
14. For most of the setup tasks, you can leave your laptop connected directly to the
management server, or you can move to a more-comfortable location and
reconnect to the management server over the customer network:
If you are going to move your connection (as described later), go to step 15.
If you are going to remain connected directly for the entire setup procedure,
read the following information, and then perform any desired steps:
The output from some commands you will type is too long to read in a
standard-size PuTTY session window. One solution is to extend the
scrolling capability. To do this, select Window in the Category list, and set
Lines of scrollback (shown in Figure 24) to 20000.
Set to 20000
scrollback
Figure 24
37
Setting up VPLEX
Another solution to long displays is to save the session output to a log file.
Over the course of setting up a dual-engine or quad-engine cluster, the log
file can grow quite large. However, if you want to create a log file anyway:
i. Select Logging in the PuTTY Configuration Category list.
ii. Under Session logging, select All session output.
iii. Type a name in the Log file name field (shown in Figure 25).
Note: Over the course of setting up a dual-engine or quad-engine cluster, the log
file can grow quite large.
Note: The log file saves data from the current PuTTY session. Any time you open a
new session, you must repeat this step if you want to log the output for that session.
putty_log
Figure 25
15. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
16. Click Open to establish a connection to the management server, and log in with
username service and password Mi@Dim7T.
Note the following:
If the connection fails to open, ping 128.221.252.3 (the LAN port on your
laptop) to verify intercomponent communication. If the ping is successful,
ping 128.221.252.2 (the service port on the management server).
The default Linux shell prompt for username service is
service@<current_working_directory>.
17. From the Linux shell prompt, type the following command to verify that the
VPlexcli process started:
sudo /etc/init.d/VPlexElementManager status
38
Setting up VPLEX
where <hostname> is a name for the management server, such as Cluster1. (The
cluster is assigned a Cluster ID of 1 automatically.) This name replaces the default
name (service) in the shell prompt the next time you open an SSH session to the
management server.
19. From the shell prompt, type the following command, to verify connectivy to
director 1-A:
ping 128.221.252.35 -c 5
Quad-engine
cluster
Engine 4:
Director 4-B
Director 4-A
128.221.252.42
128.221.252.41
Engine 3:
Director 3-B
Director 3-A
128.221.252.40
128.221.252.39
Engine 2:
Director 2-B
Director 2-A
128.221.252.38
128.221.252.37
Engine 1:
Director 1-B
Director 1-A
128.221.252.36
128.221.252.35
Dual-engine
cluster
Single-engine
cluster
Figure 26
39
Setting up VPLEX
25. Disconnect the red service cable from your laptop and the management server,
and return the cable to its location on the inside of the cabinet.
26. Prepare for connection to the customer network:
If you are connecting your laptop to the network, set it to an IP address on the
network.
If you are connecting from a different PC (or similar device):
a. Restore your laptop to the settings it had before you began this procedure
(including re-enabling any network cards).
b. Verify that PuTTY or other Telnet/SSH client is installed on the device, and
install it if necessary.
c. Set the device to an IP address on the customer network.
27. Launch PuTTY.exe, and configure PuTTY as shown in Figure 27.
Address of public
Ethernet port on
management server
(default)
Session name,
such as: VPLEX
(default)
PuTTY5
Figure 27
28. The output from some commands you will type is too long to read in a
standard-size PuTTY session window:
One solution is to extend the scrolling capability. To do this, select Window in
the Category list, and set Lines of scrollback (shown in Figure 28) to 20000.
40
Setting up VPLEX
Set to 20000
scrollback
Figure 28
Another solution to long displays is to save the session output to a log file.
Over the course of setting up a dual-engine or quad-engine cluster, the log file
can grow quite large. If you want to create a log file anyway:
a. Select Logging in the PuTTY Configuration Category list.
b. Under Session logging, select All session output.
c. Type a name in the Log file name field (shown in Figure 29).
Note: The log file saves data from the current PuTTY session. Any time you open a
new session, you must repeat this step if you want to log the output for that session.
putty_log
Figure 29
41
Setting up VPLEX
29. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
30. Click Open, and log in to the management server with username service and
password Mi@Dim7T.
31. Type the following command to connect to the VPlexcli:
telnet localhost 49500
When you see the message Configuration Complete, proceed to Task 3: Verify the
product version.
In the output, confirm that the Product Version matches the required version in the
VPLEX release notes.
Verify that the output lists all directors in the cluster, and that all directors show the
following:
Operational status: ok
Communication status: ok
42
/engines/engine-1-1/directors:
Name
Director ID
------------- ----------------------------- ----------------director-1-1-A 0x000000003ca00183
director-1-1-B 0x000000003cb00183
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
/engines/engine-1-2/directors:
Name
Director ID
------------- ----------------------------- ----------------director-1-2-A 0x000000003ca00165
director-1-2-B 0x000000003cb00165
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
Setting up VPLEX
Task 5: Open a
second PuTTY
session
Some setup tasks require executing commands from the VPlexcli prompt, and others
require excuting commands from the Linux shell prompt. You will find it easier to
leave two PuTTY sessions open on the management server, and use the applicable
session for each task.
For ease of identification, tasks that you perform in one of these sessions include one
of the following icons:
.
To open a new Linux shell session, leave the VPlexcli session open, and follow these
steps:
1. Launch PuTTY.exe.
2. Configure PuTTY as shown in Figure 27 on page 40, and then click Save.
3. Click Open, and log in to the management server with username service and
password Mi@Dim7T.
1.
2. Use vi or a similar editor to open the file, and edit the file if necessary to ensure
that it contains the following contents (with no spaces or newlines):
director_count=<n>
Task 8: Expose
back-end storage
1.
From the VPlexcli prompt, type the following command to list all
director port WWNs (identified in the command output as addresses):
ll /engines/**/ports -p
43
Setting up VPLEX
Note: The VPlexcli commands ll (two lowercase Ls) and ls support a paging option that
allows you to display long output one page at a time.
To enable paging, include the option -p at the end of the command.
To scroll from one page of the command output to the next, press the spacebar.
To exit paging mode while you are viewing paged output, type q and press Enter.
(The remainder of the output appears in its entirety.)
Note: As the ll command shows, if you want a single command to affect multiple entities
(for example, directors or ports), you sometimes can use an asterisk as a wildcard (called
command globbing). The VPLEX CLI Guide provides more detail.
2. Zone the storage to the port WWNs of the VPLEX back-end ports, following the
recommendations in the Implementation and Planning Best Practices for EMC
VPLEX Technical Notes, which are available on Powerlink.
Task 9: Verify
storage volume
availability
1.
Output example:
Storage-VolumeSummary
-----------Health
1.
Vendor
(no tier)
----------------out-of-date
storage-volumes
unhealthy
EMC
0
452
0
452
Use
unclaimed
452
Capacity
total
271G
2. From the ll output, identify the two back-end storage volumes (78 GB minimum)
to be used for the meta-volume and a mirror of the meta-volume.
Note: EMC recommends that the meta-volume and mirror be on separate arrays.
44
Setting up VPLEX
3.
where:
<meta-volume_name> is a name you are assigning to the meta-volume.
<storage-volume_1_name> is the VPD (Virtual Private Database) name of the
meta-volume.
<storage-volume_2_name> is the VPD name of the mirror.
The mirror can consist of multiple storage volumes (which will become a
RAID 1), in which case you would include each additional volume, separated
by commas. For example:
<storage-volume_1_name>,<storage-volume_2_name>,<storage-volume_3_name>
4.
In the output, verify that the attribute active shows a value of true.
5.
The meta-volume takes time to resynchronize with the mirror. Wait until the
output from cluster status shows an operational-status of ok before proceeding.
Also verify that there are no health indications in the output.
Output example:
Cluster cluster-1
operational-status:
ok
transitioning-indications:
transitioning-progress:
health-state:
ok
health-indications:
From the VPlexcli prompt, type the following command to list the
attributes of engine 1, and make a note of the top-level-assembly in the output. This
is the TLA (which is the same for each engine in a cluster) that you will need to
specify in a later task:
ll /engines/engine-1-1
45
Setting up VPLEX
From the shell prompt, type the following command from the
directory /opt/emc/connectemc:
Note: The -d and -i arguments are optional.
./connectemcConfig -n <notification_type> -e <server_address>
-s <sender_email_address> -r <recipient_email_address> [-d
<connection_description>] [-i <attach_call_home_data>]
where:
<notification type> is one of the following types:
Type
Use
Primary
Failover
onSuccess
onFailure
onSuccess/onFailure
onAllFailure
EMC recommends that you configure the following notifications, in the order
listed:
Primary notification
At least one Failover notification (for high availability)
Any necessary customer notifications
<server_address> is the address or hostname of the email server.
If ESRS is being used for the primary notifcation, enter the ESRS gateway
server IP address.
If ESRS is being used for a secondary notifcation, enter a local SMTP IP
address.
If ESRS is not being used, Call Home uses the customer email server.
Note: EMC personnel can find ESRS documentation on the GTS website, in the
Application Download area, under EMC Secure Remote Support Gateway >
Implementation > Install Procedure.
46
Setting up VPLEX
2.
Output example:
Checking EMC VPlex connectemc status
3.
Running
Note: In the current VPLEX release, the Management Console is called Element Manager
in some commands and output.
Output example:
Checking VPlex Element Manager status
4.
Running
Output example:
Checking VPlex EmaAdaptor status
5.
Running
47
Setting up VPLEX
6.
Value
----true
7.
8. If Call Home was set up to notify a customer, the email should arrive at the
customers email address. If Call Home was set up to notify EMC, the email
should arrive at the EMC Support Center.
The VPlexcli includes a set of commands (described in the EMC VPLEX CLI Guide) to
collect system configuration for the EMC System Reporting database (SYR) .
To configure a weekly SYR job, type the commands in the following
steps:
Note: If you are configuring the second cluster in a Metro-Plex, set the same schedule that is set
on the first cluster.
where:
<day> is 0 for Sunday, 1 for Monday, and so on, through 6 for Saturday.
<hours> is 0 (midnight) 23 (11:00 PM), based on a 24-hour clock.
<minutes> is 0 59.
2. Confirm the schedule:
scheduleSYR list
Output example:
SYR data collection job is currently scheduled at:
Day of Week: 0 (Sunday=0, Monday=1,...Saturday=6)
Hours: 3
Minutes: 0
48
Setting up VPLEX
This collects system configuration data for SYR, produces a file named
<V-Plex_TLA>_Config_<time_stamp>.zip, and sends the file to the address
specified in the Call Home setup. Ensure that the SYR team verifies this file.
1.
As the following output example shows, all storage volumes must have dual
paths, and must be visible from all VPLEX directors:
Summary
Cluster cluster-1
This cluster has 0 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 0 storage-volumes which are not visible from all
directors
If all storage volumes do not have dual paths, the output might be similar to the
following:
Storage volumes that are dead or unreachable:
Cluster
--------cluster-1
Director
-------------director-1-1-B
Summary
Cluster cluster-1
This cluster has 1 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 1 storage-volumes which are not visible from all
directors
2.
For each director in the cluster, type the following command, and
verify that the output shows all expected connections:
connectivity director <director_name>
Port WWN
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
Port
A2-FC00
A3-FC00
A3-FC00
A2-FC00
A2-FC00
A3-FC00
A3-FC00
A2-FC00
49
Setting up VPLEX
1.
Note that the asterisks are wildcards that replace the following entries:
First wildcard All engines in the cluster.
Second wildcard All directors in the cluster.
Third wildcard All A-side ports and all B-side ports. The [0-1] that follows
this wildcard limits the selections to all port numbers that start with A0, A1,
B0, or B1.
Fourth wildcard All ports whose numbers start with A0-FC, A1-FC, B0-FC,
or B1-FC.
2.
Type the following command to verify that all I/O ports are up:
ll /engines/**/ports -p
Address
-----------------0x5000144250014d00
0x5000144250014d01
0x5000144250014d02
0x5000144250014d03
0x5000144250014d10
0x5000144250014d11
0x5000144250014d12
0x5000144250014d13
0x5000144250014d20
0x5000144250014d21
0x5000144250014d22
0x5000144250014d23
Role
--------front-end
front-end
front-end
front-end
front-end
front-end
front-end
front-end
back-end
back-end
back-end
back-end
Port Status
----------no-link
no-link
no-link
no-link
no-link
no-link
no-link
no-link
up
up
up
up
If you are configuring a VPLEX Local implementation, verify that all wan-com
ports (A4-FC02, A4-FC03, B4-FC02, and B4-FC03) in each engine have a Port
Status of down. If one or more wan-com ports have any other status, type the
following command:
set /**/*4-FC0[2-3]::enabled false
50
ESRS must be deployed by EMC personnel. If ESRS will be used in the VPLEX
implementation, EMC personnel should follow the procedure described in the ESRS
documentation on the GTS website, in the Application Download area, under EMC
Secure Remote Support Gateway > Implementation > Install Procedure.
Setting up VPLEX
Zone the front-end and back-end I/O ports as recommended in the following:
Implementation and Planning Best Practices for EMC VPLEX Technical Notes,
available on Powerlink
Use a VPlexcli session or the VPLEX Management Console GUI to perform this task.
Recommendations:
If a storage volume is equal to or larger than the desired virtual volume capacity,
create an extent the size of the desired virtual volume.
If a storage volume is smaller than the desired virtual volume capacity, create a
single extent per storage volume, and then use devices to concatenate or stripe
these extents into a larger device.
Create one view for each set of hosts that require access to the same storage.
You can use the VPlexcli or the Management Console GUI to expose storage. The
basic steps are the same for either:
To use the VPlexcli, proceed to If you are using the VPlexcli on page 52.
Figure 30
51
Setting up VPLEX
Note: The VPLEX CLI guide contains details on using VPlexcli commands.
where:
<storage_volume_name> is an unique name you are assigning to the volume.
<storage_volume_ID> is the ID (for example, Symm2206_2ADF, or
clar2065_LUN537) in the output from step 1.
Note: If you want to claim multiple storage volumes, separate the volume IDs with
commas or use the command storage-volume claimingwizard, as described in the
VPLEX CLI Guide.
In you do not include any arguments, this command creates an extent the size of
the storage volume. However, you can create multiple smaller extents (up to 128
per storage volume).
Note: Extents are named and claimed automatically.
4. Type the following command to verify that the extent was created:
ls /clusters/cluster-1/storage-elements/extents
extent_Symm1707_1002_1
extent_Symm1707_1006_1
extent_Symm1707_100A_1
extent_Symm1707_100E_1
extent_Symm1707_1012_1
extent_Symm1707_1016_1
extent_Symm1707_101A_1
extent_Symm1707_101E_1
extent_Symm1707_1022_1
extent_Symm1707_1026_1
extent_Symm1707_102A_1
extent_Symm1707_102E_1
extent_Symm1707_1032_1
extent_Symm1707_1036_1
extent_Symm1707_103A_1
extent_Symm1707_1003_1
extent_Symm1707_1007_1
extent_Symm1707_100B_1
extent_Symm1707_100F_1
extent_Symm1707_1013_1
extent_Symm1707_1017_1
extent_Symm1707_101B_1
extent_Symm1707_101F_1
extent_Symm1707_1023_1
extent_Symm1707_1027_1
extent_Symm1707_102B_1
extent_Symm1707_102F_1
extent_Symm1707_1033_1
extent_Symm1707_1037_1
extent_Symm1707_103B_1
52
Setting up VPLEX
port-wwn
-----------------0x10000000c866fff2
node-wwn
-----------------0x20000000c866fff2
type
------default
habc4101_pt1
0x10000000c866fff3
0x20000000c866fff3
default
habc4102_pt0
0x10000000c866ffec
0x20000000c866ffec
default
habc4102_pt1
0x10000000c866ffed
0x20000000c866ffed
default
9. Type the following commands to create and export the storage view:
cd /clusters/cluster-1/exports
storage-view create -c /clusters/cluster-1/ -n <storage_view_name>
-p <VPLEX_port>,<VPLEX_port>
10. Type the following commands to add each initiator port to the storage view:
cd /clusters/cluster-1/exports
storage-view addinitiatorport -v <storage_view_name> -i
<initiator_port>,<initiator_port>
11. Type the following commands to add the virtual volume(s) to the storage view:
cd /clusters/cluster-1/exports
storage-view addvirtualvolume -v <storage_view_name> -o
<virtual-volume_name>
Note that for any virtual volumes you add after the first one, you must add -f to
the end of the command. For example:
storage-view addvirtualvolume -f -v <storage_view_name> -i
<virtual-volume_name>,<virtual-volume_name>
53
Setting up VPLEX
Note: Including -f forces the virtual volumes to be added to the view even if the specified
LUNs are in use, the view is live, the virtual volumes are already assigned to another view,
or there are problems determining the view's state.
12. Type the following commands to verify the status of the storage view:
cd /clusters/cluster-1/exports/storage-views
export storage-view summary
Output example:
View health summary(cluster-1):
view name
health-state exported volumes
ports
----------VMware_view
view1
view2
view3
view4
----4
4
4
4
4
-----------healthy
healthy
healthy
healthy
healthy
---------------2
1311
10
11
10
registered
initiators
---------4
480
2
2
2
54
After provisioning and exporting storage, perform a rescan from the host to verify
that the exported virtual volume is visible.
Setting up VPLEX
Before you can log in to a VPLEX management server over the customer network,
you must set the address on the management servers public Ethernet port:
Note: The first cluster you set up will be designated cluster 1, and the second cluster will be
cluster 2. Several tasks in this guide refer to the management servers on cluster 1 and cluster 2
as management server 1 and management server 2.
Note: The tasks in this document use PuTTY for logging in to the management server.
However, you can use any similar Telnet/SSH client.
1. Remove the red service cable that hangs inside the cabinet of the first VPLEX
cluster you are setting up, and use the cable to connect your laptop to the clusters
management server as shown in Figure 31.
Laptop
Service port
Mgmt-B port
Figure 31
Mgmt-A port
zep-009b
2. On your laptops toolbar, select Start > Settings > Control Panel > Network
Connections (double-click).
3. If any public network connection is enabled on your laptop, right-click the
network card and select Disable.
4. On the Network Connections window, right-click Local Area Connection and
select Properties.
5. On the Local Area Connection Properties window, double-click Internet
Protocol (TCP/IP) under This connection uses the following items.
6. On the Internet Protocol (TCP/IP) Properties window:
a. If your laptop is configured with specific addresses, make a note of them, to
ensure that you reset the configuration correctly later.
b. Select Use the following IP addresses, and type the following values:
IP address:
128.221.252.3
Subnet mask:
255.255.255.224
Default gateway: 128.221.252.2
55
Setting up VPLEX
7. Click OK on the Internet Protocol (TCP/IP) Properties window, and again on the
Local Area Connection Properties window.
8. Close the Network Connections window.
9. Open a Command Prompt window (Start > Programs > Accessories > Command
Prompt), and type the following command to confirm the settings:
ipconfig
Output example:
Ethernet adapter Local Area Connection:
Connection-specific
IP Address. . . . .
Subnet Mask . . . .
Default Gateway . .
DNS
. .
. .
. .
Suffix
. . . .
. . . .
. . . .
.
.
.
.
:
:
:
:
eng.emc.com
128.221.252.3
255.255.255.224
128.221.252.2
10. Type the following command to verify that you can reach the management
servers service port:
ping 128.221.252.2
Type:
128.221.252.2
(default)
Type session name:
VPLEX_service
(default)
PuTTY
Figure 32
13. Select SSH in the Category list, and ensure that Preferred SSH protocol version
is set to 2, as shown in Figure 33.
56
Setting up VPLEX
SSH protocol
version
putty_ssh
Figure 33
14. For most of the setup tasks, you can leave your laptop connected directly to the
management server, or you can move to a more-comfortable location and
reconnect to the management server over the customer network:
If you are going to move your connection (as described later), go to step 15.
If you are going to remain connected directly for the entire setup procedure on
this management server, follow these steps:
Note: You will perform the same steps on the second management server later.
a. Select Window in the Category list, and set Lines of scrollback (shown in
Figure 34 on page 58) to 20000. (This allows you to read long output in a
standard-size PuTTY session window.)
57
Setting up VPLEX
Set to 20000
scrollback
Figure 34
Note: The log file saves data from the current PuTTY session. Any time you open a
new session, you must repeat this step if you want to log the output for that session.
58
Setting up VPLEX
putty_log
Figure 35
15. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
16. On the PuTTY Configuration window, click Open to establish a connection to the
management server, and log in with username service and password Mi@Dim7T.
Note the following:
If the connection fails to open, ping 128.221.252.3 (the LAN port on your
laptop) to verify intercomponent communication. If the ping is successful,
ping 128.221.252.2 (the service port on the management server).
The default Linux shell prompt for username service is
service@<current_working_directory>.
17. From the Linux shell prompt, type the following command, and verify that the
VPlexcli process started:
sudo /etc/init.d/VPlexElementManager status
where <hostname> is a name for the management server, such as Cluster1. This
name replaces the default name (service) in the shell prompt the next time you
open a PuTTY session to the management server
19. From the shell prompt, type the following command, to verify connectivy to
director 1-A:
ping 128.221.252.35 -c 5
59
Setting up VPLEX
20. Repeat step 19 once for each director in the cluster, substituting the applicable
address shown in Figure 36.
Quad-engine
cluster
Engine 4:
Director 4-B
Director 4-A
128.221.252.42
128.221.252.41
Engine 3:
Director 3-B
Director 3-A
128.221.252.40
128.221.252.39
Engine 2:
Director 2-B
Director 2-A
128.221.252.38
128.221.252.37
Engine 1:
Director 1-B
Director 1-A
128.221.252.36
128.221.252.35
Dual-engine
cluster
Single-engine
cluster
Zep-056
Figure 36
25. Disconnect the red service cable from the management server and hang it back in
the cabinet.
26. If the clusters are located at different sites, go to step 35.
If the clusters are at the same location, use the red service cable in the second
clusters cabinet to connect to the service port on the management server, and
then proceed to step 27.
27. Launch PuTTY, and load the VPLEX_Service session you saved earlier in this
task.
28. Click Open, and log in with username service and password Mi@Dim7T.
60
Setting up VPLEX
29. From the Linux shell prompt, type the following command, and verify that the
VPlexcli process started:
sudo /etc/init.d/VPlexElementManager status
32. From the VPlexcli prompt, type the following command to configure the IP
address of the management servers public Ethernet port:
management-server set-ip -i <IP_address>:<netmask> -g <gateway> eth3
33. Type the following command at the VPlexcli prompt, and again at the Linux shell
prompt:
exit
34. Disconnect the red service cable from the management server and hang it back in
the cabinet.
35. Prepare for connection to the customer network:
If you are connecting your laptop to the network, set it to an IP address on
the network.
If you are connecting from a different PC (or similar device):
a. Restore your laptop to the settings it had before you began this
procedure (including re-enabling any network cards).
b. Verify that PuTTY or other Telnet/SSH client is installed on the device,
and install it if necessary.
c. Set the device to an IP address on the customer network.
61
Setting up VPLEX
VPLEX service
(default)
PuTTY2
Figure 37
37. Select Window in the Category list, and set Lines of scrollback (shown in
Figure 38) to 20000. (This allows you to read long output in a standard-size
PuTTY session window.)
Set to 20000
scrollback
Figure 38
Setting up VPLEX
Note: Over the course of setting up a dual-engine or quad-engine cluster, the log file can
grow quite large.
Note: The log file saves data from the current PuTTY session. Any time you open a new
session, you must repeat this step if you want to log the output for that session.
putty_log
Figure 39
39. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
40. Configure a new PuTTY session as shown in Figure 40.
Type the address of the
public Ethernet port on
the management server
in the second cluster you
are configuring
(default)
VPLEX service
VPLEX Cluster 1
(default)
PuTTY3
Figure 40
63
Setting up VPLEX
The setup wizard performs several tasks to set up a VPLEX implementation, based on
a series of prompts for user-entered information. Refer to the EMC VPLEX
Pre-installation Configuration Worksheet as you respond to the prompts.
From the VPlexcli prompt, type the following command, and follow the steps in the
automated interview and configuration procedure:
configuration system-setup
When you see the message Configuration Complete, proceed to Task 3: Verify the
product version.
In the output, confirm that the Product Version matches the required version in the
VPLEX release notes.
Verify that the output lists all directors in the cluster, and that all directors show the
following:
Operational status: ok
Communication status: ok
64
/engines/engine-1-1/directors:
Name
Director ID
------------- ----------------------------- ----------------director-1-1-A 0x000000003ca00183
director-1-1-B 0x000000003cb00183
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
/engines/engine-1-2/directors:
Name
Director ID
------------- ----------------------------- ----------------director-1-2-A 0x000000003ca00165
director-1-2-B 0x000000003cb00165
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
Setting up VPLEX
Task 5: Open a
second PuTTY
session
Some setup tasks require executing commands from the VPlexcli prompt, and others
require excuting commands from the Linux shell prompt. You will find it easier to
leave two PuTTY sessions open on the management server, and use the applicable
session for each task.
To open a new Linux shell session, leave the VPlexcli session open, and follow these
steps:
1. Launch PuTTY.exe.
2. Configure PuTTY as described in step 37 on page 62 through step 39 on page 63.
3. Click Open, and log in to the management server with username service and
password Mi@Dim7T.
For ease of identification, many of the remaining tasks in this section include one of
the following icons to identify the associated session:
.
1.
2. Use vi or a similar editor to open the file, and edit the file if necessary to ensure
that it contains the following contents (with no spaces or newlines):
director_count=<n>
2. Ping switch B:
ping 128.221.253.34
Task 8: Expose
back-end storage
1.
From the VPlexcli prompt, type the following command to list all
director port WWNs (identified in the command output as addresses):
ll /engines/**/ports -p
65
Setting up VPLEX
Note: The VPlexcli commands ll (two lowercase Ls) and ls support a paging option that
allows you to display long output one page at a time.
To enable paging, include the option -p at the end of the command.
To scroll from one page of the command output to the next, press the spacebar.
To exit paging mode while you are viewing paged output, type q and press Enter.
(The remainder of the output appears in its entirety.)
Note: As the ll command shows, if you want a single command to affect multiple entities
(for example, directors or ports), you sometimes can use an asterisk as a wildcard (called
command globbing). The VPLEX CLI Guide provides more detail.
2. Zone the storage to the port WWNs of the VPLEX back-end ports, following the
recommendations in the Implementation and Planning Best Practices for EMC
VPLEX Technical Notes, which are available on Powerlink.
Task 9: Verify
storage volume
availability
1.
Output example:
Storage-VolumeSummary
-----------Health
1.
Vendor
(no tier)
----------------out-of-date
storage-volumes
unhealthy
EMC
0
452
0
452
Use
unclaimed
452
Capacity
total
271G
From the ll output, identify the two back-end storage volumes (78 GB minimum)
to be used for the meta-volume and a mirror of the meta-volume.
Note: EMC recommends that the meta-volume and mirror be on separate arrays.
66
Setting up VPLEX
2.
where:
<meta-volume_name> is a name you are assigning to the meta-volume.
<storage-volume_1_name> is the VPD (Virtual Private Database) name of the
meta-volume.
<storage-volume_2_name> is the VPD name of the mirror.
The mirror can consist of multiple storage volumes (which will become a
RAID 1), in which case you would include each additional volume, separated
by commas. For example:
<storage-volume_1_name>,<storage-volume_2_name>,<storage-volume_3_name>
3.
In the output, verify that the attribute active shows a value of true.
4.
The meta-volume takes time to resynchronize with the mirror. Wait until the
output from cluster status shows an operational-status of ok before proceeding.
Also verify that there are no health indications in the output.
Output example:
Cluster cluster-1
operational-status:
ok
transitioning-indications:
transitioning-progress:
health-state:
ok
health-indications:
From the VPlexcli prompt, type the following command to list the
attributes of engine 1, and make a note of the top-level-assembly in the output. This
is the TLA (which is the same for each engine in a cluster) that you will need to
specify in a later task:
ll /engines/engine-1-1
67
Setting up VPLEX
From the shell prompt, type the following command from the
directory /opt/emc/connectemc:
Note: The -d and -i arguments are optional.
./connectemcConfig -n <notification_type> -e <server_address>
-s <sender_email_address> -r <recipient_email_address> [-d
<connection_description>] [-i <attach_call_home_data>]
where:
<notification type> is one of the following types:
Type
Use
Primary
Failover
onSuccess
onFailure
onSuccess/onFailure
onAllFailure
EMC recommends that you configure the following notifications, in the order
listed:
Primary notification
At least one Failover notification (for high availability)
Any necessary customer notifications
<server_address> is the address or hostname of the email server.
If ESRS is being used for the primary notifcation, enter the ESRS gateway
server IP address.
If ESRS is being used for a secondary notifcation, enter a local SMTP IP
address.
If ESRS is not being used, Call Home uses the customer email server.
Note: EMC personnel can find ESRS documentation on the GTS website, in the
Application Download area, under EMC Secure Remote Support Gateway >
Implementation > Install Procedure.
68
Setting up VPLEX
2.
Output example:
Checking EMC VPlex connectemc status
3.
Running
Note: In the current VPLEX release, the Management Console is called Element Manager
in some commands and output.
Output example:
Checking VPlex Element Manager status
4.
Running
Output example:
Checking VPlex EmaAdaptor status
5.
Running
69
Setting up VPLEX
6.
Value
----true
7.
8. If Call Home was set up to notify a customer, the email should arrive at the
customers email address. If Call Home was set up to notify EMC, the email
should arrive at the EMC Support Center.
The VPlexcli includes a set of commands (described in the EMC VPLEX CLI Guide) to
collect system configuration for the EMC System Reporting database (SYR) .
To configure a weekly SYR job, type the commands in the following
steps:
1. Set the schedule:
scheduleSYR add -d <day> -t <hours> -m <minutes>
where:
<day> is 0 for Sunday, 1 for Monday, and so on, through 6 for Saturday.
<hours> is 0 (midnight) 23 (11:00 PM), based on a 24-hour clock.
<minutes> is 0 59.
2. Confirm the schedule:
scheduleSYR list
Output example:
SYR data collection job is currently scheduled at:
Day of Week: 0 (Sunday=0, Monday=1,...Saturday=6)
Hours: 3
Minutes: 0
This collects system configuration data for SYR, produces a file named
<V-Plex_TLA>_Config_<time_stamp>.zip, and sends the file to the address
specified in the Call Home setup. Ensure that the SYR team verifies this file.
70
Setting up VPLEX
1.
As the following output example shows, all storage volumes must have dual
paths, and must be visible from all VPLEX directors:
Summary
Cluster cluster-1
This cluster has 0 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 0 storage-volumes which are not visible from all
directors
If all storage volumes do not have dual paths, the output might be similar to the
following:
Storage volumes that are dead or unreachable:
Cluster
--------cluster-1
Director
-------------director-1-1-B
Summary
Cluster cluster-1
This cluster has 1 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 1 storage-volumes which are not visible from all
directors
2.
For each director in the cluster, type the following command, and
verify that the output shows all expected connections:
connectivity director <director_name>
Port WWN
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
Port
A2-FC00
A3-FC00
A3-FC00
A2-FC00
A2-FC00
A3-FC00
A3-FC00
A2-FC00
71
Setting up VPLEX
Type the following command to enable all front-end ports in the cluster:
set /engines/*/directors/*/hardware/ports/*[0-1]-FC*::enabled true
Note that the asterisks are wildcards that replace the following entries:
First wildcard All engines in the cluster.
Second wildcard All directors in the cluster.
Third wildcard All A-side ports and all B-side ports. The [0-1] that follows
this wildcard limits the selections to all port numbers that start with A0, A1,
B0, or B1.
Fourth wildcard All ports whose numbers start with A0-FC, A1-FC, B0-FC,
or B1-FC.
1. Did you set the public IP address on both management servers in Task 1 on
page 55?
Note: As a reminder, the only scenario in which you would have done this is if both
clusters are at the same location and you connected to the cluster over the customer
network after you set the IP address on the management server.
If yes:
a. Type the following command at the VPlexcli prompt, and again at the
Linux shell prompt:
exit
2. Remove the red service cable that hangs inside cluster 2s cabinet, and use the
cable to connect your laptop to the clusters management server as shown in
Figure 41.
72
Setting up VPLEX
Laptop
Service port
Mgmt-B port
Figure 41
Mgmt-A port
zep-009b
3. On your laptops toolbar, select Start > Settings > Control Panel > Network
Connections (double-click).
4. If any public network connection is enabled on your laptop, right-click the
network card and select Disable.
5. On the Network Connections window, right-click Local Area Connection and
select Properties.
6. On the Local Area Connection Properties window, double-click Internet
Protocol (TCP/IP) under This connection uses the following items.
7. On the Internet Protocol (TCP/IP) Properties window:
a. If your laptop is configured with specific addresses, make a note of them, to
ensure that you reset the configuration correctly later.
b. Select Use the following IP addresses, and type the following values:
IP address:
128.221.252.3
Subnet mask:
255.255.255.224
Default gateway: 128.221.252.2
8. Click OK on the Internet Protocol (TCP/IP) Properties window, and again on the
Local Area Connection Properties window.
9. Close the Network Connections window.
10. Open a Command Prompt window (Start > Programs > Accessories > Command
Prompt), and type the following command to confirm the settings:
ipconfig
11. Type the following command to verify that you can reach the management
servers service port:
ping 128.221.252.2
73
Setting up VPLEX
Type:
128.221.252.2
(default)
Type session name:
VPLEX_service
(default)
PuTTY
Figure 42
14. Select SSH in the Category list, and ensure that Preferred SSH protocol version
is set to 2, as shown in Figure 43.
SSH protocol
version
putty_ssh
Figure 43
74
Setting up VPLEX
15. For most of the setup tasks, you can leave your laptop connected directly to the
management server, or you can move to a more-comfortable location and
reconnect to the management server over the customer network:
If you are going to move your connection (as described later), go to step 16.
If you are going to remain connected directly for the entire setup procedure,
follow these steps:
a. Select Window in the Category list, and set Lines of scrollback (shown in
Figure 44) to 20000. (This allows you to read long output in a standard-size
PuTTY session window.)
Set to 20000
scrollback
Figure 44
Note: The log file saves data from the current PuTTY session. Any time you open a
new session, you must repeat this step if you want to log the output for that session.
75
Setting up VPLEX
putty_log
Figure 45
16. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
17. On the PuTTY Configuration window, click Open to establish a connection to the
management server, and log in with username service and password Mi@Dim7T.
Note the following:
If the connection fails to open, ping 128.221.252.3 (the LAN port on your
laptop) to verify intercomponent communication. If the ping is successful,
ping 128.221.252.2 (the service port on the management server).
The default Linux shell prompt for username service is
service@<current_working_directory>.
18. From the Linux shell prompt, type the following command to verify that the
VPlexcli process started:
sudo /etc/init.d/VPlexElementManager status
where <hostname> is a name for the management server, such as Cluster2. This
name replaces the default name (service) in the shell prompt the next time you
open a PuTTY session to the management server
20. From the shell prompt, type the following command, to verify connectivy to
director 1-A:
ping 128.221.252.67 -c 5
76
Setting up VPLEX
21. Repeat step 19 once for each director in the cluster, substituting the applicable
address shown in Figure 36.
Quad-engine
cluster
Engine 4:
Director 4-B
Director 4-A
128.221.252.74
128.221.252.73
Engine 3:
Director 3-B
Director 3-A
128.221.252.72
128.221.252.71
Engine 2:
Director 2-B
Director 2-A
128.221.252.70
128.221.252.69
Engine 1:
Director 1-B
Director 1-A
128.221.252.68
128.221.252.67
Dual-engine
cluster
Single-engine
cluster
Zep-056a
Figure 46
26. Disconnect the red service cable from the management server and hang it back in
the cabinet.
27. Prepare for connection to the customer network:
If you are connecting your laptop to the network, set it to an IP address on the
network.
If you are connecting from a different device:
a. Restore your laptop to the settings it had before you began this procedure
(including re-enabling any network cards).
77
Setting up VPLEX
b. Verify that PuTTY or other Telnet/SSH client is installed on the device, and
install it if necessary.
c. Set the device to an IP address on the customer network.
28. Launch PuTTY.exe, and configure PuTTY as shown in Figure 47.
Type the address of the
public Ethernet port on
the management server
in the second cluster you
are configuring
(default)
VPLEX service
VPLEX Cluster 1
(default)
PuTTY3
Figure 47
29. Select Window in the Category list, and set Lines of scrollback (shown in
Figure 48) to 20000. (This allows you to read long output in a standard-size
PuTTY session window.)
Set to 20000
scrollback
Figure 48
78
Setting up VPLEX
Note: The log file saves data from the current PuTTY session. Any time you open a new
session, you must repeat this step if you want to log the output for that session.
putty_log
Figure 49
31. Select Session in the Category list, and click Save in the PuTTY Configuration
window to save the settings for the session you configured.
32. Click Open, and log in to the management server with username service and
password Mi@Dim7T.
33. Type the following command to connect to the VPlexcli:
telnet localhost 49500
79
Setting up VPLEX
The setup wizard performs several tasks to set up a VPLEX implementation, based on
a series of prompts for user-entered information. Refer to the EMC VPLEX
Pre-installation Configuration Worksheet as you respond to the prompts
From the VPlexcli prompt, type the following command, and follow the steps in the
automated interview and configuration procedure:
configuration system-setup
When you see the message Configuration Complete, proceed to Task 18: Verify the
product version.
Type the following command from the VPlexcli prompt, to display the software
versions:
version -a
In the output, confirm that the Product Version matches the required version in the
VPLEX release notes.
Verify that the output lists all directors in the cluster, and that all directors show the
following:
Operational status: ok
Communication status: ok
80
/engines/engine-2-1/directors:
Name
Director ID
------------- ----------------------------- ----------------director-2-1-A 0x000000003ca00183
director-2-1-B 0x000000003cb00183
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
/engines/engine-2-2/directors:
Name
Director ID
------------- ----------------------------- ----------------director-2-2-A 0x000000003ca00165
director-2-2-B 0x000000003cb00165
Cluster
ID
------1
1
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
Setting up VPLEX
Some setup tasks require executing commands from the VPlexcli prompt, and others
require excuting commands from the Linux shell prompt. You will find it easier to
leave two PuTTY sessions open on the management server, and use the applicable
session for each task.
To open a new Linux shell session, leave the VPlexcli session open, and follow step 28
on page 78 through step 32 on page 79.
For ease of identification, many of the remaining tasks include one of the following
icons:
.
1.
2. Use vi or a similar editor to open the file, and edit the file if necessary to ensure
that it contains the following contents (with no spaces or newlines):
director_count=<n>
2. Ping switch B:
ping 128.221.253.66
1.
From the VPlexcli prompt, type the following command to list all
director port WWNs (identified in the command output as addresses):
ll /engines/**/ports -p
2. Zone the storage to the port WWNs of the VPLEX back-end ports, following the
recommendations in the Implementation and Planning Best Practices for EMC
VPLEX Technical Notes, which are available on Powerlink.
1.
81
Setting up VPLEX
Output example:
Storage-VolumeSummary
-----------Health
1.
Vendor
(no tier)
----------------out-of-date
storage-volumes
unhealthy
EMC
0
452
0
452
Use
unclaimed
452
Capacity
total
271G
From the ll output, identify the two back-end storage volumes (78 GB minimum)
to be used for the meta-volume and a mirror of the meta-volume.
Note: EMC recommends that the meta-volume and mirror be on separate arrays.
2.
where:
<meta-volume_name> is a name you are assigning to the meta-volume.
<storage-volume_1_name> is the VPD (Virtual Private Database) name of the
meta-volume.
<storage-volume_2_name> is the VPD name of the mirror.
The mirror can consist of multiple storage volumes (which will become a
RAID 1), in which case you would include each additional volume, separated
by commas. For example:
<storage-volume_1_name>,<storage-volume_2_name>,<storage-volume_3_name>
3.
In the output, verify that the attribute active shows a value of true.
82
Setting up VPLEX
4.
The meta-volume takes time to resynchronize with the mirror. Wait until the
output from cluster status shows an operational-status of ok before proceeding.
Also verify that there are no health indications in the output.
Output example:
Cluster cluster-2
operational-status:
ok
transitioning-indications:
transitioning-progress:
health-state:
ok
health-indications:
From the VPlexcli prompt, type the following command to list the
attributes of engine 1, and make a note of the top-level-assembly in the output. This
is the TLA (which is the same for each engine in a cluster) that you will need to
specify in a later task:
ll /engines/engine-2-1
From the shell prompt, type the following command from the
directory /opt/emc/connectemc:
Note: The -d and -i arguments are optional.
./connectemcConfig -n <notification_type> -e <server_address>
-s <sender_email_address> -r <recipient_email_address> [-d
<connection_description>] [-i <attach_call_home_data>]
where:
<notification type> is one of the following types:
Type
Use
Primary
Failover
onSuccess
onFailure
onSuccess/onFailure
onAllFailure
EMC recommends that you configure the following notifications, in the order
listed:
Primary notification
At least one Failover notification (for high availability)
Any necessary customer notifications
83
Setting up VPLEX
2.
Output example:
Checking EMC VPlex connectemc status
3.
Running
84
Setting up VPLEX
Note: In the current VPLEX release, the Management Console is called Element Manager
in some commands and output.
Output example:
Checking VPlex Element Manager status
4.
Running
Output example:
Checking VPlex EmaAdaptor status
5.
Running
7.
Value
----true
8. If Call Home was set up to notify a customer, the email should arrive at the
customers email address. If Call Home was set up to notify EMC, the email
should arrive at the EMC Support Center.
The VPlexcli includes a set of commands (described in the EMC VPLEX CLI Guide) to
collect system configuration for the EMC System Reporting database (SYR) .
To configure a weekly SYR job, type the commands in the following
steps. Set the same schedule that is set on cluster 1.
1. Set the schedule:
scheduleSYR add -d <day> -t <hours> -m <minutes>
85
Setting up VPLEX
where:
<day> is 0 for Sunday, 1 for Monday, and so on, through 6 for Saturday.
<hours> is 0 (midnight) 23 (11:00 PM), based on a 24-hour clock.
<minutes> is 0 59.
2. Confirm the schedule:
scheduleSYR list
Output example:
SYR data collection job is currently scheduled at:
Day of Week: 0 (Sunday=0, Monday=1,...Saturday=6)
Hours: 3
Minutes: 0
This collects system configuration data for SYR, produces a file named
<V-Plex_TLA>_Config_<time_stamp>.zip, and sends the file to the address
specified in the Call Home setup. Ensure that the SYR team verifies this file.
1.
As the following output example shows, all storage volumes must have dual
paths, and must be visible from all VPLEX directors:
Summary
Cluster cluster-2
This cluster has 0 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 0 storage-volumes which are not visible from all
directors
If all storage volumes do not have dual paths, the output might be similar to the
following:
Storage volumes that are dead or unreachable:
Cluster
Dead or Unreachable Storage Volumes
--------- ---------------------------------------cluster-2 VPD83T3:60000970000192601426533032463334
Storage volumes that are not visible at all directors:
Cluster
Director
Storage Volumes Director Cannot See
--------- -------------- ---------------------------------------cluster-2 director-2-1-B VPD83T3:60000970000192601426533032463334
Summary
Cluster cluster-2
This cluster has 1 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 1 storage-volumes which are not visible from all
directors
86
Setting up VPLEX
2.
For each director in the cluster, type the following command, and
verify that the output shows all expected connections:
connectivity director <director_name>
Port WWN
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0009000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
0x0093000000000000
Port
A2-FC00
A3-FC00
A3-FC00
A2-FC00
A2-FC00
A3-FC00
A3-FC00
A2-FC00
Type the following command to enable all front-end ports in the cluster:
set /engines/*/directors/*/hardware/ports/*[0-1]-FC*::enabled true
Note that the asterisks are wildcards that replace the following entries:
First wildcard All engines in the cluster.
Second wildcard All directors in the cluster.
Third wildcard All A-side ports and all B-side ports. The [0-1] that follows
this wildcard limits the selections to all port numbers that start with A0, A1,
B0, or B1.
Fourth wildcard All ports whose numbers start with A0-FC, A1-FC, B0-FC,
or B1-FC.
1.
2.
In the output, verify that all ports numbered A4-FC02, A4-FC03, B4-FC02, or
B4-FC03 have a port status of up.
87
Setting up VPLEX
ESRS must be deployed by EMC personnel. If ESRS will be used in the VPLEX
implementation, EMC personnel should follow the procedure described in the ESRS
documentation on the GTS website, in the Application Download area, under EMC
Secure Remote Support Gateway > Implementation > Install Procedure.
Several of the remaining tasks require moving back and forth between VPlexcli
sessions to both clusters. You will find it easier to leave two sessions open.
Some tasks include one of the following icons to identify the associated PuTTY
session:
Management server 1:
Management server 2:
1.
Output example:
Tue Feb 16 19:07:58 UTC 2010
2.
3.
88
Setting up VPLEX
1.
2. Type the following command to display all director ports in the cluster, and verify
that all ports numbered A4-FC02, A4-FC03, B4-FC02, or B4-FC03 have a port
status of up:
ll /engines/engine-1-*/**/ports
Implementation and Planning Best Practices for EMC VPLEX Technical Notes,
available on Powerlink
1.
1.
89
Setting up VPLEX
2. Type the following command to verify that cluster 2 is connected to all of the
directors in cluster 1:
ls /engines/engine-1-*/directors
1.
2.
In the output, verify that both clusters have the same Island ID and an operational
status of ok.
Output example:
Clusters:
Name
Cluster ID Connected Expelled
Operational Status
Health
------------------ --------- ----------------------------------------------------------------------------------------------------------cluster-1
1
true
false
ok
ok
cluster-2
2
true
false
ok
ok
Islands:
Island ID
--------1
Clusters
-------------------cluster-1, cluster-2
You must create a logging volume on each VPLEX Metro cluster. The volume must be
large enough to contain one bit for every page of distributed storage space
(approximately 10 GB of logging volume space for every 320 TB of distributed
devices).
The logging volumes experience much I/O during and after link outages. The
recommended best practice is to stripe each logging volume across many disks for
speed, and also to have a mirror (on another fast disk), because this is important data.
CAUTION
If a logging volume does not exist, every link failure could cause a full
resynchronization of every distributed device in the system.
90
Setting up VPLEX
To create the logging volumes, you first must claim the storage volumes that will be
used, and create extents from those volumes:
1.
VPlexcli-2
From the VPlexcli prompt on management server 2, type the
following commands to list the storage volumes:
cd /clusters/cluster-<Cluster_ID>/storage-elements/storage-volumes
ll
2.
VPlexcli-2
Type the following command to claim the storage volume(s) that
you will use for the logging volume:
storage-volume claim -n <storage_volume_name> -d
<storage_volume_ID>
where:
<storage_volume_name> is an unique name you are assigning to the volume.
<storage_volume_ID> is the ID (for example, Symm2206_2ADF, or
clar2065_LUN537) in the output from step 1.
Note: If you want to claim multiple storage volumes, separate the volume IDs with
commas or use the command storage-volume claimingwizard, as described in the
VPLEX CLI Guide.
3.
VPlexcli-2
Repeat this step for each extent you want to create on the cluster.
4.
VPlexcli-2
cd /clusters/cluster-<Cluster_ID>/system-volumes
logging-volume create -n <name> -g <raid_geometry> -e <extent_1> ,
<extent_2> [, <extent_3> ... <extent_n>] -d <stripe_depth>
where:
<name> is a name for the logging volume.
<raid_geometry> is the RAID type: raid-0 or raid-1.
<extent_> identifies the name(s) of the extent(s) that make(s) up the logging
volume.
-d <stripe_depth> is required only if <geometry> is raid-0.
The stripe depth must be:
Greater than zero, but no greater than the number of blocks of the smallest
element of the RAID 0 device being created
A multiple of 4000 bytes
Note: A depth of 32 means that 128 K (32 x 4 K) is written to the first disk, the next
128 K is written to the next disk, and so on.
91
Setting up VPLEX
5.
1.
Cluster
ID
------2
2
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
/engines/engine-2-2/directors:
Name
Director ID
------------- ----------------------------- ----------------director-2-2-A 0x000000003ca00165
director-2-2-B 0x000000003cb00165
Cluster
ID
------2
2
Commissioned
----------------------true
true
Operational
Status
----------ok
ok
Communication
Status
-----------ok
ok
2.
1.
In the output, confirm that operational-status is ok and that there are no health
indications.
2.
In the output, verify that both clusters have the same Island ID and an operational
status of ok.
In the output, confirm that all storage volumes in the cluster have dual paths, and are
visible from all VPLEX directors.
Output example:
Summary
Cluster cluster-2
This cluster has 0 storage-volumes which are dead or unreachable
This cluster has 0 storage-volumes which do not have dual paths
This cluster has 0 storage-volumes which are not visible from all
directors
92
Setting up VPLEX
In the output, confirm that all storage volumes in the cluster have dual paths, and are
visible from all VPLEX directors.
Use a VPlexcli session or the VPLEX Management Console GUI to perform this task.
Recommendations:
If a storage volume is equal to or larger than the desired virtual volume capacity,
create an extent the size of the desired virtual volume.
If a storage volume is smaller than the desired virtual volume capacity, create a
single extent per storage volume, and then use devices to concatenate or stripe
these extents into a larger device.
Create one view for each set of hosts that require access to the same storage.
You can use the VPlexcli or the Management Console GUI to expose storage. The
basic steps are the same for either:
3. To begin provisioning and exporting storage, click Provision Storage on the main
screen of the VPLEX Management Console GUI (shown in Figure 50).
Figure 50
93
Setting up VPLEX
Note: The VPLEX CLI guide contains details on using VPlexcli commands.
where:
<storage_volume_name> is an unique name you are assigning to the volume.
<storage_volume_ID> is the ID (for example, Symm2206_2ADF, or
clar2065_LUN537) in the output from step 1.
Note: If you want to claim multiple storage volumes, separate the volume IDs with
commas or use the command storage-volume claimingwizard, as described in the
VPLEX CLI Guide.
In you do not include any arguments, this command creates an extent the size of
the storage volume. However, you can create multiple smaller extents (up to 128
per storage volume).
Note: Extents are automatically named and claimed.
4. Type the following command to verify that the extent was created:
ls /clusters/cluster-<Cluster_ID>/storage-elements/extents
extent_Symm1707_1002_1
extent_Symm1707_1006_1
extent_Symm1707_100A_1
extent_Symm1707_100E_1
extent_Symm1707_1012_1
extent_Symm1707_1016_1
extent_Symm1707_101A_1
extent_Symm1707_101E_1
extent_Symm1707_1022_1
extent_Symm1707_1026_1
extent_Symm1707_102A_1
extent_Symm1707_102E_1
extent_Symm1707_1032_1
extent_Symm1707_1036_1
extent_Symm1707_103A_1
extent_Symm1707_1003_1
extent_Symm1707_1007_1
extent_Symm1707_100B_1
extent_Symm1707_100F_1
extent_Symm1707_1013_1
extent_Symm1707_1017_1
extent_Symm1707_101B_1
extent_Symm1707_101F_1
extent_Symm1707_1023_1
extent_Symm1707_1027_1
extent_Symm1707_102B_1
extent_Symm1707_102F_1
extent_Symm1707_1033_1
extent_Symm1707_1037_1
extent_Symm1707_103B_1
94
Setting up VPLEX
extent_Symm1707_1002_1
extent_Symm1707_1006_1
extent_Symm1707_100A_1
extent_Symm1707_100E_1
extent_Symm1707_1012_1
extent_Symm1707_1016_1
extent_Symm1707_101A_1
extent_Symm1707_101E_1
extent_Symm1707_1022_1
extent_Symm1707_1026_1
extent_Symm1707_102A_1
extent_Symm1707_102E_1
extent_Symm1707_1032_1
extent_Symm1707_1036_1
extent_Symm1707_103A_1
extent_Symm1707_1003_1
extent_Symm1707_1007_1
extent_Symm1707_100B_1
extent_Symm1707_100F_1
extent_Symm1707_1013_1
extent_Symm1707_1017_1
extent_Symm1707_101B_1
extent_Symm1707_101F_1
extent_Symm1707_1023_1
extent_Symm1707_1027_1
extent_Symm1707_102B_1
extent_Symm1707_102F_1
extent_Symm1707_1033_1
extent_Symm1707_1037_1
extent_Symm1707_103B_1
9. Type the following commands to create and export the storage view:
cd /clusters/cluster-<Cluster_ID>/exports
storage-view create -c /clusters/cluster-<Cluster_ID>/ -n
<storage_view_name> -p <VPLEX_port>,<VPLEX_port>
10. Type the following commands to add each initiator port to the storage view:
cd /clusters/cluster-<Cluster_ID>/exports
storage-view addinitiatorport -v <storage_view_name> -i
<initiator_port>,<initiator_port>
11. Type the following commands to add the virtual volume(s) to the storage view:
cd /clusters/cluster-<Cluster_ID>/exports
storage-view addvirtualvolume -v <storage_view_name> -o
<virtual-volume_name>
Note that for any virtual volumes you add after the first one, you must add -f to
the end of the command. For example:
storage-view addvirtualvolume -f -v <storage_view_name> -i
<virtual-volume_name>,<virtual-volume_name>
Note: Including -f forces the virtual volumes to be added to the view even if the specified
LUNs are in use, the view is live, the virtual volumes are already assigned to another view,
or there are problems determining the view's state.
95
Setting up VPLEX
12. Type the following commands to verify the status of the storage view:
cd /clusters/cluster-<Cluster_ID>/exports/storage-views
export storage-view summary
Output example:
View health summary(cluster-1):
view name
health-state exported volumes
ports
----------VMware_view
view1
view2
view3
view4
----4
4
4
4
4
-----------healthy
healthy
healthy
healthy
healthy
---------------2
1311
10
11
10
registered
initiators
---------4
480
2
2
2
96
After provisioning and exporting storage, perform a rescan from the host to verify
that the exported virtual volume is visible.
A
Reference
This appendix contains miscellaneous information you will need during the
installation. The appendix also contains tables to record other information you might
need.
Reference
97
Reference
128.221.253.42
128.221.253.41
Engine 4:
Director 4B
Director 4A
128.221.252.42
128.221.252.41
Engine 3:
Director 3B
Director 3A
128.221.253.40
128.221.253.39
Engine 3:
Director 3B
Director 3A
128.221.252.40
128.221.252.39
FC switch B 128.221.253.34
Service port
128.221.252.2
Mgt B port
128.221.253.33
Mgt A port
128.221.252.33
Management server
Engine 2:
Director 2B
Director 2A
128.221.253.38
128.221.253.37
Engine 2:
Director 2B
Director 2A
128.221.252.38
128.221.252.37
Engine 1:
Director 1B
Director 1A
128.221.253.36
128.221.253.35
Engine 1:
Director 1B
Director 1A
128.221.252.36
128.221.252.35
Zep-028_1
Figure 51
98
Reference
Cluster IP Seed = 2
Enclosure IDs = engine numbers
Management network B addresses
Engine 4:
Director 4B
Director 4A
128.221.253.74
128.221.253.73
Engine 4:
Director 4B
Director 4A
128.221.252.74
128.221.252.73
Engine 3:
Director 3B
Director 3A
128.221.253.72
128.221.253.71
Engine 3:
Director 3B
Director 3A
128.221.252.72
128.221.252.71
FC switch B 128.221.253.66
Service port
128.221.252.2
Mgt B port
128.221.253.65
Mgt A port
128.221.252.65
Management server
Engine 2:
Director 2B
Director 2A
128.221.253.70
128.221.253.69
Engine 2:
Director 2B
Director 2A
128.221.252.70
128.221.252.69
Engine 1:
Director 1B
Director 1A
128.221.253.68
128.221.253.67
Engine 1:
Director 1B
Director 1A
128.221.252.68
128.221.252.67
Zep-028_2
Figure 52
99
Reference
Internal cabling
The figures in this section show the various cabling inside a VPLEX cabinet. The
figures are included here for reference only, because the cables are installed before the
unit ships from EMC.
The figures include cable lengths and EMC part numbers, for use in verifying or
replacing a cable.
This section includes the following figures:
Cluster size
Cable type
Figure
Quad-engine
Ethernet
Serial
Fibre Channel
AC power
Ethernet
Serial
Fibre Channel
AC power
Ethernet
Serial
Fibre Channel
AC power
Dual-engine
Single-engine
100
Reference
Quad-engine cluster
038-003-676, 71 in.
038-003-887, 48 in.
038-003-679, 20 in.
038-003-678, 20 in.
Engine 4
Engine 3
038-003-678, 20 in.
Management server
038-003-677, 37 in.
038-003-679, 20 in.
038-003-685, 37 in.
038-003-677, 37 in.
038-003-685, 37 in.
Engine 2
Engine 1
Zep-05_el
Figure 53
Internal cabling
101
Reference
Engine 4
038-003-682
12 in.
038-003-682
12 in.
Engine 3
038-003-682
12 in.
038-003-682
12 in.
UPS B
UPS A
038-003-894
40 in.
038-003-894
40 in.
Engine 2
038-003-682
12 in.
038-003-682
12 in.
Engine1
038-003-682
12 in.
038-003-682
12 in.
Zep-05_sl
Figure 54
102
Reference
Engine 4
Engine 3
038-003-738, 79 in.
(all 16 cables)
Engine 2
Engine 1
Zep-05_fl
Figure 55
Internal cabling
103
Reference
ON
I
ON
I
O
OFF
O
OFF
Engine 4
SPS 4
ON
I
ON
I
O
OFF
O
OFF
Engine 3
ON
I
ON
I
O
OFF
O
OFF
SPS 3
Engine 2
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
SPS 2
O
OFF
O
OFF
ON
I
ON
I
Engine 1
SPS 1
Zep-005_acl
Figure 56
104
Reference
Dual-engine cluster
038-003-676, 71 in.
Management server
038-003-677, 37 in.
038-003-679, 20 in.
038-003-678, 20 in.
038-003-887, 48 in.
038-003-676, 71 in.
Engine 2
Engine 1
Zep-05_em
Figure 57
Internal cabling
105
Reference
UPS B
UPS A
038-003-894
40 in.
038-003-894
40 in.
Engine 2
038-003-682
12 in.
038-003-682
12 in.
Engine1
038-003-682
12 in.
038-003-682
12 in.
Zep-05_sm
Figure 58
106
Reference
Engine 2
Engine 1
Zep-05_fm
Figure 59
Internal cabling
107
Reference
I
O
I
I
O
SPS 2
I
O
I
O
Engine 1
SPS 1
Zep-005_acm
Figure 60
108
Reference
Single-engine cluster
038-003-677, 37 in.
038-003-676, 71 in.
Management server
Engine 1
Zep-005_es
Figure 61
Engine 1
038-003-682
12 in.
038-003-682
12 in.
Zep-005_ss
Figure 62
Engine 1
038-003-737,
39 in. (2 cables)
Zep-005_fs
Figure 63
Internal cabling
109
Reference
Management server
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
Engine 1
SPS
Zep-005_acs
Figure 64
110
Reference
Notes:
The upper PDPs are installed, but
are not used in V-Plex configurations.
Outlet group
circuit breaker
OG
ON
I
O
OFF
O
OFF
Unused PDP
OG
Power distribution
unit (PDU)
OG
ON
I
OG
ON
I
ON
I
O
OFF
O
OFF
OG
OG
ON
I
ON
I
O
OFF
O
OFF
Outlet groups
O
OFF
O
OFF
ON
I
ON
I
OG
OG
OG
OG
OG
OG
O
OFF
Power distribution
panel (PDP)
O
OFF
ON
I
ON
I
O
OFF
O
OFF
ON
I
ON
I
To 30 A, 220 VAC
power source 2
Rear view
To 30 A, 220 VAC
power source 1
Zep-012
Figure 65
111
Reference
112
Index
ls 39
acclimating to environment 25
addresses of hardware components 98
C
cabling, internal
dual-engine cluster 105
quad-engine cluster 101
single-engine cluster 109
Call Home, configuring 46
circuit breakers, PDU 30
cluster, defined 16
command globbing 39
connectivity, back-end, verifying 49
E
Element Manager 47
ESRS, setting up 50
EZ-Setup wizard 42
meta-volume
creating 44
described 20
P
paging mode, VPlexcli, exiting 39
paging through VPlexcli command output 39
password, service account 38
port roles 26
ports, enabling
front-end 50
WAN COM 87
power connections to PDUs 111
power switches, PDP 30
PuTTY, sessions, multiple 43
S
F
Fibre Channel ports 26
G
GbE ports 26
H
hardware required 20
I
I/O ports 26
IP addresses 98
setup wizard 42
software required 21
storage, exporting 51
SYR, configuring 48
T
TLA, determining 45
tools needed 21
U
username, service account 38
W
wizard, EZ-Setup 42
LEDs
management server 32
SPS 31
UPS 31
ll 39
113
Index
114