Docu48727 - Unisphere-1 3 0 1 0015-Release-Notes
Docu48727 - Unisphere-1 3 0 1 0015-Release-Notes
1.3.0.1.0015
Release Notes
P/N 300-015-045
REV 01
August 16, 2013
1
Revision history
Revision history
Product description
Unisphere
® ®
EMC Unisphere 1.3 is a unified interface for configuring and
®
managing the EMC VNX series of hardware platforms. It can also
be used to manage CX4, CX3, CX™, or AX4-5 series storage
systems. These systems provide a single platform for data access
for File and Block.
The Unisphere 1.3 software, either Unisphere Server or Unisphere
Client (off-array version), may be installed on Windows™ Server
2003, Windows Server 2008, Windows Server 2012, Windows Vista,
Windows XP, Windows 7, and Windows 8 Servers.
Unisphere Analyzer
EMC Unisphere Analyzer gathers storage system performance
statistics and presents them in various customizable charts. These
charts can help users to identify existing and potential bottlenecks
of data access, and provide information for load balancing, capacity
planning, as well as cost-benefit analysis.
Analyzer collects and manages performance data of SPs, LUNs,
disks, and so on. Performance data can be archived to create
historical record file which allows offline analysis. Analyzer keeps
a limited amount of performance data by default; when the
maximum amount is reached, the latest data will replace the oldest.
Analyzer has a web-based user interface which allows users to
securely monitor the performance of EMC VNX and EMC
®
CLARiiON storage systems either locally in the same LAN or
remotely over the Internet. The web-based user interface is
downloaded and run by the web browser.
Analyzer supports EMC VNX, CX4, CX3, CX™, or AX4-5 series
storage systems.
Table 2. Host Software, CLI and Utilities environment and system requirements
Operating Minimum Minimum software requirements
system hardware
requirements
Solaris Sun system OS Solaris 9 SPARC
capable of Solaris 10 SPARC
accepting PCI, Solaris 10 x64 (Opteron)
SBus, or cPCI
HBAs with at least Oracle Solaris 11 Express x86
30 MB of free disk Oracle Solaris SPARC
space and 64 MB Oracle Solaris 11 x86
of RAM Other libstdc++.so.6 (part of package SUNWgccruntime)
is needed.
Windows Intel-compatible OS Windows Server 2003 (SP2) x86, x64, IA64
PC with 64 MB of Windows Server 2008 x86, x64, IA64
RAM and at least Windows Server 2008 R2 x64, IA64
130 MB of free
disk space Windows Server 2008 R2 (SP1) x64, IA64
Windows Server 2008 Server Core x64, IA64
Windows Server 2008 R2 Server Core x64, IA64
Windows Server 2012 x64
Windows Vista x86, x64
Windows 7 x64
Windows 8 x64
Others For iSCSI support, the iSCSI initiator driver should
be version 2.4 or later.
AIX IBM AIX OS IBM AIX 6.1 (P5, P6 and P7)
workstation with at IBM AIX 7.1 (P6 and P7)
least 64 MB of
RAM and 40 MB of Others The AIX runtime environment should be at level 8
free disk space or later.
The Unisphere Host Agent listens on port 6389. Ensure the firewall
(if enabled) allows both inbound and outbound communication on
this port. The machine running Unisphere Host Agent requires an
Version 1.3.0.1.0015
Unisphere
Symmetrix VMAX support on VNX Gateway systems
Link and launch to Unisphere for VMAX from VNX Unisphere for
Gateway systems
Unisphere, running on VG10 and VG50 Gateway systems only,
now includes links for directly launching Unisphere for VMAX.
Clients connect to Unisphere for VMAX only by using HTTPS.
In order to use this Link and launch feature, you must first register
Unisphere for VMAX. Credentials include a client ID and a client
password that you must pre-configure on the UVMAX before
registering in Unisphere for VNX. In order to pre-configure these
credentials, you must have a UVMAX Administrator, Manager, or
Monitor role.
You register UVMAX credentials once for a selected Symmetrix
array, and those credentials are used throughout the session
whenever a UVMAX window is accessed. You can register multiple
Symmetrix arrays per Unisphere for VNX session.
Only one system can be powered off at a time. Once the power off
process is triggered, it does not stop until the process completes
and cannot be cancelled or rolled back, even if errors occur during
the power off process. You must manually disconnect the power
cables after the power off process completes.
Pool LUN default type is now Thin LUN
If the Thin Provisioning enabler is installed, when you create a new
pool LUN in Unisphere, a thin LUN will now be created by default.
If you want to create a pool LUN that is not thin, you must clear the
Thin checkbox during LUN creation, whether you are using the
LUN creation window or the LUN Provisioning wizard. When
running a LUN creation CLI command, if you do not explicitly
state the type of pool LUN to create, then a thick pool LUN will be
created.
Unisphere Analyzer
Multi-Core Cache (MCC) Enhancements
Unisphere Analyzer performance archive files now display a new
performance characteristic for SPs: SP Cache Flush
Request/s. This statistic indicates the number of requests made
when writing (synchronizing) data from SP cache to disks to free
up cache pages for incoming write data. The lower the value, the
fewer requests SP cache is making to flush out data or free up
pages for incoming writes, meaning the system is utilizing cache
more effectively.
Fixed problems
Version 1.3.0.1.0015
There are no fixed problems as this is the initial release.
Unisphere
Brief description Symptom details Solution (or
workaround)
Platforms: More LUNs cannot be In the Storage > LUNs In the case where pool
created in the Storage > table, more LUNs LUNs reach the limit,
Severity: LUNs table by clicking cannot be created by but RAID group LUNs
Create when either the clicking Create when do not, create RAID
Low group LUNs in
pool LUNs or RAID group either the pool LUNs or
LUNs have reached the RAID group LUNs Storage > Storage
Tracking limit. reach the limit. A Pools > RAID
Number: popup warning Groups.
574138 message will appear
and then the create In the case where
LUN dialog will close. RAID group LUNs
reach the limit, but
pool LUNs do not,
create pool LUNs in
Storage > Storage
Pools > Pools.
Platforms: After issuing the getlog If there are many log There are two steps to
All command where there are entries in the event this workaround:
many event log log, the getlog 1. Ensure the
entries,CIMOM may command may output network
Severity:
restart and thus stop a lot of information, bandwidth
Low
service temporarily. which can take a long between the host
time. If getlog which issued the
command takes over getlog command
Tracking half an hour, a dump and the target
Exists in versions:
1.3.0.1.0015
Platforms: LUN compression fails if if the system is Delete another LUN,
All the system is configured configured with then compress the
with the maximum number maximum LUNs, target LUN.
of LUNs. enabling the LUN
Severity: compression will fail.
Low Exists in versions:
1.3.0.1.0015
Tracking
Number:
490729
Platforms: Error when pinging/tracing The following error Do not ping/trace route
All the route SP management message will display SP management IP
IP from an iSCSI data port when pinging/tracing from an iSCSI data
in the UI. the route SP port; it is not
Severity: management IP from recommended.
Medium an iSCSI data port in
Unisphere UI: Exists in versions:
Tracking Ping/traceroute
Number: to peer SP or 1.3.0.1.0015
management
522202
workstation from
iSCSI port
disrupts
management
communication and
is not permitted.
Exists in versions:
1.3.0.1.0015
Platforms: Two hosts with the same In the UI, two hosts There are two
All name are displayed in the with the same host workarounds for this
host list in the UI. name can be seen in issue:
the host list. In the CLI,
Severity: the storagegroup -
Medium 1. In order to connect
connecthost
back the logged-
command cannot
out paths:
Tracking assign all the paths
that belong to the a. First, connect
Number: back the down
548376 same host to the target
storage group. paths (those
paths that are
logged-out, but
still regiestered
on the array).
b. Restart the Host
Agent service
on the host.
2. If the down paths
were removed as
planned and will
no longer be used,
unregister the
Exists in versions:
1.3.0.1.0015
Platforms: Unisphere and the CLI do The UI and CLI do not Perform a full poll.
VNX8000 not accurately display the correctly display the
60-drive Disk Array 60-drive enclosure Exists in versions:
Enclosure (DAE) information when
Severity: information after replacing replacing a 25-drive 1.3.0.1.0015
Medium a 25-drive DAE with a 60- DAE with a 60-drive
drive DAE. DAE.The information
Tracking displayed may be a
Number: mix of information from
563290 the two enclosures.
Platforms: After a bus cable is moved If a bus cable is moved This is a display issue
All from one port to another, from one port to only. Restart the
faults may be seen on another, there will be management server
both the original bus and faults from both the from the setup page or
Severity: the new bus. original bus and the reboot the SP.
Medium new bus.This will
remain even after Exists in versions:
Tracking moving the bus cable
back to the first port. 1.3.0.1.0015
Number:
563384
Platforms: A faulted LUN does not A faulted LUN will not No workaround.
All display in the output of the be shown in the
faults –list command faults -list output Exists in versions:
unless it is associated with unless it is associated
Severity: 1.3.0.1.0015
a system feature. with a system feature
Medium such as a storage
group, mirror, snap,
Tracking and so on.
Number:
564905
Exists in versions:
1.3.0.1.0015
Platforms: Creation of FAST Cache When both the storage Destroy the FAST
All may fail if created at the pool and FAST Cache Cache and recreate it.
same time as the storage are created
pool. simultaneously, FAST Exists in versions:
Severity: Cache may fail to be
Critical created. 1.3.0.1.0015
Tracking
Number:
561400
Platforms: Two Windows 2008 iSCSI When running Log out of Unisphere
Windows hosts connected at the Unisphere, after two and log in again.
Server 2008 same time do not display Windows 2008 iSCSI
under Host > Host List. hosts are connected to Exists in versions:
an array and rebooted,
Severity: although they are 1.3.0.1.0015
Medium shown as logged in
and registered under
Tracking Hosts > Initiators, the
Number: following warning is
526482 shown on the status
column.
The initiator is
not fully
connected to the
Storage System.
As a result they do not
show under Hosts >
Host List.
Exists in versions:
1.3.0.1.0015
Platforms: Unisphere hangs after When connected to a Workaround:
VNX5800 attaching LUNs to a system with a large 1. After adding the
storage group on a configuration (for LUNs, wait
VNX5800 with a large example, with approximately 30
Severity: configuration. around1000 LUNs)
Critical seconds before
that also has I/O closing the
running through the confirmation dialog
Tracking Unisphere UI, 2. Restart
Number: Unisphere may hang theUnisphere UI
574400 when attaching a LUN to proceed.
to a storage group.
Unisphere Analyzer
Brief description Symptom details Solution (or
workaround)
Platforms: Unisphere may crash For systems with large Increase Java memory
All when using Analyzer on a configurations, the by:
system with a large Unisphere UI may 1. Go the Start menu
configuration (hundreds of crash if there are in Windows.
Severity: LUNs). hundreds of LUNs.
Critical 2. Select Settings >
Control Panel.
3. Double-click the
Tracking
Java Plugin icon.
Number:
4. Select the
572930
Advanced tab.
5. In the Java Runtime
Parameters, type
Xmx1024m.
Exists in versions:
1.3.0.1.0015
Tracking
Number:
572417
Platforms: Pool creation failed on a When attempting to No workaround. This is
VNX5400 VNX5400. create a storage pool, expected behavior
the following pool when the system limits
status message may are exceeded. The
Severity: be received if the storage pool can be
Low system LUN limits destroyed.
have been reached:
Tracking Current
Exists in versions:
Number: Operation:
Creating Current 1.3.0.1.0015
573776
Operation State:
Failed Current
Operation Status:
Illegal unit
number Current
Operation Percent
Completed: 24
This message
indicates that system
configuration has
reached it maximum
configuration for FLU's
and the pool cannot be
created.
Platforms: Removal of an iSNS target Although the Server No workaround. The
Windows is reported by the Server Utility cannot remove Server Utility cannot
Utility as successful when an iSNS target, the be used to remove an
the operation was actually iSNS target still can be iSNS target.
Severity: unsuccessful. selected for removal.
Medium The remove operation Exists in versions:
will then be reported
Tracking as completing 1.3.0.1.0015
Number: successfully, even
573772 though it was
unsuccessful and the
target is still present.
Exists in versions:
1.3.0.1.0015
Platforms: Unisphere Host Agent Unisphere Host Agent The hostname-
All Linux and cannot be started on an cannot be started on ipaddress mapping
UNIX AsianUx host. AsianUx host. An error should be added to
platforms message such as the /etc/hosts. Host
following in Agent requires the
/var/log/agent.lo hostname-ipaddress
Severity: g is displayed: mapping for a network
Critical
mm/dd/yyyy connection.
16:43:56 Agent
Tracking Main -- Net or
File event. Err: Exists in versions:
Number:
Local 1.3.0.1.0015
547526
hostname/address
mapping unknown;
see installation
notes.
Platforms: On a Windows 2012 On Windows Server Install Microsoft
Windows server, Windows STOP 2012 without February Windows 8 and
Server 2012 errors can occur when 2013 cumulative Windows Server 2012
running Server Utility 1.3 updates installed, February 2013
or Host Agent 1.3. Windows STOP errors cumulative updates.
Exists in versions:
1.3.0.1.0015
Platforms: When configuring an Server Utility may fail If multiple network
Windows iSCSI connection, the to connect if the server adapters are
Server Utilty may fail to has serveral network configured on the
connect if the wrong adapters, not all of server, EMC
Severity: network adapter is them are reachable to recommends that the
Critical selected. each other, and the default adapter is
wrong network adapter selected while using
Tracking is selected when the Server Utility to
Number: configuring an iSCSI create an iSCSI
537021 connection. The failure connection.
message may also Alternatively, ensure
take a while to appear. that the right network
adapter is selected.
Otherwise, the
selected subnet may
not be able to access
the target. Make sure
that the selected
network adapter can
access the target
configured on array.
Exists in versions:
1.3.0.1.0015
2620:0:170:1d58::
af5:581f
Name:
dempster-cs
Port:
80
Technical notes
Unisphere
Legacy systems
Celerra
Unisphere supports legacy systems running Celerra Manager with
NAS version 6.0.36.4 or later.
Navisphere
Unisphere supports legacy systems running Navisphere® release
6.24 (release 6.23 for AX) or later.
Change
Error code Severity in Previous Value New Value
Could not save the profile.
Possible reasons could be
an invalid profile format, or
a Storage Processor
internal error. Ensure that
the profile format is
correct. If this does not fix
the problem, gather
diagnostic data and
contact your service
0x4141 Warn eventText Could not update profile. provider.
Change
Error code Severity in Previous Value New Value
The Compression
operation could not be
performed on this LUN The Compression
because the LUN has operation could not be
been selected for performed on this LUN
migration into a dedup because deduplication is
0x71658228 Error eventText pool. in progress for this LUN.
The snapshot mount
point cannot be added to
the consistency group
because it is attached to
a different snapshot than
the other members of
that consistency group. Unable to add snapshot
Detach the snapshot mount point to
then add the snapshot consistency group
mount point to the because it is attached to
0x716D8029 Error eventText consistency group. an individual snapshot.
severity Warn Error
The snapshot's primary
LUN is enabling or
disabling deduplication.
The snapshot will be The snapshot will be
deleted at the end of destroyed when
migration as part of enabling/disabling
enabling/disbaling deduplication is complete.
0x8919 Error eventText dedup. LUN: %1
Snapshot cannot be
restored because it
Cannot restore snapshot contains a LUN that is
while deduplication is enabling or disabling
0x8920 Error eventText enabling/disabling. deduplication. LUN: %1
Analyzer
Configuration
Unisphere off-array tool does not support Analyzer on-array
functions
Analyzer in an off-array Unisphere server does NOT support the
functions designed to work in on-array. For example, the
Performance Data Logging dialog box is grayed out on an off-array
Unisphere server.
Configuration modifications are not saved
When the storage systems running Navisphere 6.29 or Unisphere
1.0 are in the same domain as storage systems running earlier
Navisphere releases, you will see an Analyzer Customize dialog
box. If you log in to one storage system running Navisphere 6.29 or
Unisphere 1.0 as a global user, modify configurations and save
them and then log in to another storage system running an earlier
version of Navisphere as that same global user to check these
configurations, the values are not consistent with what you
modified in the first storage system.
Create at least one LUN on the storage system before using
Analyzer
If a storage system does not contain at least one LUN , menu items
related to Analyzer will be grayed-out and the generated archive
will not contain any useful data. Before using Analyzer, ensure at
least one LUN is created on the storage system.
JRE requirements
EMC highly recommends using JRE 1.7 update 17. Analyzer works
on JRE 1.6 update 15 or later. For more information about JRE
Merging archives
If the FLARE version is greater than 03.24 or 02.24, archives created
by any SP are independent. Do not merge archives from different
arrays.
Beginning with Unisphere version 1.2, when opening a merged file
from numbers of Analyzer archive files and GUI performance
charts, a prompt message Analyzer does not display all the
data points in the selected time range, some of them
are ignored is added to the status bar of the performance chart.
Do not merge archives from both SPs of the same array with time
range overlap.
Performance issues
Monitoring hundreds of objects degrades performance
Concurrently monitoring hundreds of objects in real-time may
cause the management interface to become sluggish. To improve
response time, increase the real-time polling interval or reduce the
number of objects selected to be monitored.
System with large configuration
Beginning with Unisphere 1.0, the storage system can support
thousands of LUNs and metaLUNs. Analyzer may experience
performance issues on a storage system with a very large
configuration. See the following suggestions.
To improve Analyzer performance on a storage system with a large
configuration:
If a storage system has more than 4096 objects, set the archive
interval and real-time interval to 600 seconds or more.
EMC recommends creating an archive every 12 hours so that
each archive contains 80 to 90 samples. Smaller archives are
easier to view and manipulate.
Set the JRE memory to 1024 MB or higher if working with an
archive with a large number of objects. If memory is not
adequate, you will see an out-of-memory exception, and
operations in progress will not finish.
When using the UI to open an archive, clear Initially check all
tree nodes in the archive tab in the Customize dialog box. This
will improve the response time when opening the chart. After
the chart opens, manually selecting the objects and statistics to
view also reduces response time. Do not use the select all items
option for a large configuration since this causes slow
performance.
Choosing dump provides more options for selecting object
types and attributes. The UI dump is faster than the CLI dump,
Control methods
Performance fluctuations
When you run a policy using the cruise control method to achieve
certain performance targets, QoS Manager goes through a period of
information gathering and calibrating. This information gathering
and calibrating can take some time depending on the condition of
I/O activity and policy goal. Normally, it is between several
minutes to half an hour.
During this process, QoS Manager gathers information about
Thin provisioning
I/O to thin LUNs exhibits different behaviors from I/O to thick
LUNs. The response time of I/O to thin LUNs is not as predictable
as that of I/O to thick LUNs and is more difficult for the control
algorithms in the QoS Manager to manage. When thin LUNs are in
an I/O class, you should expect it to take longer to achieve target
performance defined by cruise control policies and wider
Compression
The response time of compressed I/O is not as predictable as that
of thick LUNs and is more difficult for the control algorithms in the
QoS Manager to manage. During the initial compression operation,
I/O to compressed LUNs may have delayed response times which
impact the ability of control algorithms to maintain goals. In steady
state operations like thin LUNs, you should expect it to take longer
to achieve target performance defined by cruise control policies
and wider variations in limits policies when compressed LUNs are
included in I/O classes. Cruise control policies with narrow
tolerances are expected to fail more frequently. Wider tolerances
will deliver better results.
Cruise Control
Cruise control relies on effects that could be weakened by Thin
Provisioning, Auto-Tiering, and VNX Snapshots technology.
Therefore, a longer attaining time and a higher possibility of goal
failure may occur when running cruise control on targets with
these features. These features include Thin LUNs, Auto-Tiered
LUNs, Compression, and Snapshot Mount Points.
Viewing policies
Viewing policies with 20 or more I/O classes in the QoS Manager
chart panel can make it difficult to identify current performance
levels for individual I/O classes. This is due to the multitude of
displayed line graphs. To alleviate this problem, use the I/O class
filtering functionality in the chart configuration options.
Hover your mouse cursor over individual data points to identify
I/O class performance levels.
QoS Manager chart color combinations
Although every effort has been made to generate unique color
combinations for the lines displayed on the QoS Manager chart
panels, it may be difficult to distinguish between colors when a 256
color video mode is used. Users are suggested to use the highest
color resolution possible. Our experience suggests that True Color
video mode provides the best experience.
Copying images from clipboard
Users of X-Window systems might encounter issues when copying
an image from the clipboard. In such a case, users are suggested to
save the graph as a JPEG file.
Off-Array UI
You can use the QoS Manager off-array UI to manage the storage
systems which have QoS Manager enabled. When you use the off-
array UI to open or dump a retrieved archive file you do not need
to connect to a storage system.
IP address.
To solve this problem, you need to create a text file called
agentID.txt. This text file tells the Host Agent the IP address to use
when it communicates with the client.
The first line of this text file is the fully qualified hostname. The
second line is the IP address. If either of these lines is empty, the
default value for that parameter is used. If you want to include
comments, create an empty third line and start your comments on
the fourth line.
Following is an example of an agentID.txt file.
Hostname
128.221.41.34
Host Agent file for VNX series, CX4, CX3 and CX series, and AX4-5 series
storage systems
Creating device entries
Do not create device entries for VNX, CX4, CX3, CX, and AX4-5
storage systems in the Host Agent configuration file.
Error codes
Error codes returned by the CLI are provided in Table 5.
If you wish to use SecureCLI with these older systems you need to
install SecureCLI in "low security mode".
0x4061 The local replication role is not supported by all nodes in the
domain.
0x4062 The replication/recovery role is not supported by all nodes in the
domain.
0x4063 The replication role is not supported by all nodes in the domain..
0x5370 Security invalid command.
0x5371 Security feature not available.
0x5372 Execute error
Domain error codes
0x4300 Event directory created.
0x4301 Event domain destroyed.
0x4302 Event directory destroyed.
0x4303 Event nodes added.
0x4304 Event nodes removed.
0x4305 Event version mismatch.
0x4306 Event local system info changed.
0x4307 Event New domain created.
0x4308 Event local master.
0x4309 Event Local node demoted.
0x430a Event New master defined.
0x430b Event node part of diff domain.
0x430c Event client data destroyed.
0x4340 Event no database in PSM.
0x4341 Event no backup database.
0x4342 Event inconsistent DIR DB.
0x4343 Event internal provider error.
0x4344 Event directory creation failed.
0x53A0 Directory invalid command
0x53A1 Directory feature not available
0x53A2 Directory execute errors
Legacy support error codes
0x4400 Gateway system not found.
0x4401 System not found in directory.
0x4402 Invalid agent configuration.
0x4403 Unable to save configuration.
Certificate verification
Certificate verification does not support MD5 certificates.
Unisphere Host Agent failing or causing core dump on AIX 6.1 and
above
When executing 32-bit version of the Host Agent on 64 bit versions
of AIX 6.1 and later, the Host Agent would receive only 256MB per
each allocated shared memory segment, which is not sufficient.
Due to this, whenever the Host Agent exceeds the size, kernel
generates SIGSEGV (signal-11) which terminates the process and
causes Host Agent to fail or a core dump.
Follow these steps on 64 bit AIX 6.1and above:
1. Go to the Host Agent installation directory:
cd /usr/lpp/UNIAGENT/
AIX — Utilities
Upgrade AIX runtime environment to version 8 or later
AIX runtime environment (rte) should be at version 8 or later. By
default, AIX 5.3 comes with default runtime environment 6. You
must install xlc.rte.aix50.feb2007.ptf from following location:
http://www-
01.ibm.com/support/docview.wss?rs=2239&context=SSJT9L&dc=
D400&uid=swg24015076&loc=en_US&cs=UTF-8&lang=en
Secure CLI
Installing Secure CLI client
Follow these steps to install Secure CLI client on HPUX systems in
"Low" security mode:
1. Uninstall any currently installed version of the Secure CLI
client.
2. Run the command:
Host Agent
IA-64 Red Hat Enterprise Linux 4.0 installation
If you get errors while trying to install Unisphere packages on an
IA-64 host, install the IA-32 Execution Layer and required 32-bit
compatibility packages for runtime support for 32-bit application.
Refer to the following procedure from Red Hat:
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-
Manual/release-notes/as-itanium/
Red Hat Enterprise Linux-4 for the 64-bit Intel-Itanium2
architecture includes runtime support for 32-bit applications
through Intel's IA-32 Execution Layer.
The IA-32 Execution Layer is provided on the Extras disk for the
Intel-Itanium2 architecture. In addition, a set of 32-bit libraries and
applications are provided on a separate 32-bit Compatibility-Layer
disk.
Together, the IA-32 Execution Layer and 32-bit compatibility
packages provide a runtime environment for 32-bit applications on
the 64-bit native distribution.
To install the IA-32 Execution Layer and required 32-bit
compatibility packages, follow these steps:
1. Install Red Hat Enterprise Linux 4 for the Intel-Itanium2
architecture.
2. During the first system boot, the Additional CDs screen
prompts you to insert the Red Hat Enterprise Linux Extras CD.
Insert the CD (on which the ia32el package is located) and click
Install if you wish to run 32-bit applications.
3. When prompted, choose Misc from the available choices to
install the ia32el and ksh packages.
4. Click Details to verify the selection of packages to be installed.
At this point the ia32el package will start itself as a service.
5. After the installation of the packages completes, the system will
Auto-detect
The Host Agent configuration file should not contain any ATF,
PowerPath or standard device entries for CX, CX3, or AX storage
systems. The auto-detect feature of the remote agent configurator
does not detect CX, CX3, and AX storage systems.
Veritas Storage Foundation for Windows
If Veritas Storage Foundation for Windows is present on a
Windows Server 2003 host, you must manually register the server.
Windows — CLI
Unable to add iSCSI host or host appears multiple times in iSCSI
configurations
If you cannot add an iSCSI host to a storage group or the iSCSI host
appears multiple times in the CLI portlist command, it may be
due to incorrect handling of case in the Microsoft Initiator.
Clusters
EMC recommends that you place all shared disks (LUNs),
including the quorum disk, into the same storage group.
Documentation
EMC provides the ability to create step-by-step planning,
installation, and maintenance instructions tailored to your
environment. To create VNX customized documentation, go to:
https://mydocs.emc.com/VNX.
For the most up-to-date documentation and help go to EMC Online
Support at https://Support.EMC.com.
Documentation Updates
LUN Snapview Snapshot Number of LUNs on the selected system that have one or
sessions more user-created snapshot sessions.
Clone sources Number of LUNs on the selected system that have one or
more clones.
Mirror sources Number of LUNs on the selected system that have
(Asynchronous) MirrorView/A mirrors, including mirrors that are part of
asynchronous groups.
Mirror sources Number of source LUNs on the selected storage system
(Synchronous) that have MirrorView/S mirrors, including mirrors that are
part of synchronous groups.
SAN Copy sessions Number of LUNs on the selected storage system that have
one or more user-created SAN Copy sessions.
LUN Primary Snapshot Number of LUNs on the selected system that are used to
sources create snapshots.
Installation
For information on installing Unisphere server software, Host
software, and Utilities, refer to https://mydocs.emc.com/VNX/.
Select Install or update server software under Server tasks and
select Update Unisphere server software.
Unisphere Quality of Service Manager (QoS) and Unisphere
Analyzer software are preloaded on all storage systems with the
OE bundles. Enabler packages are usually installed when the
storage system is set up at your site. To install or update QoS in an
existing storage system, select the Software tab in the Unisphere
Service Manager (USM). Installed enablers will remain on the
system through an upgrade. You can download USM from EMC
Online Support. To use QoS or Analyzer, refer to the Unisphere
Online Help or download the EMC VNX Command Line Interface
(CLI) Reference for Block available on EMC Online Support.
You can use QoS or Analyzer in a client mode (off-array) mode to
view QoS archives. Follow the instructions in Setting Up a Unisphere
Management Station, available on EMC Online Support, to set up a
management station. You do not need to install the enabler on the
Unisphere management station in order to view QoS archives.
2. HP-UX
On HP-UX, the SecureCLI packages can be installed non-
interactively by setting the ask option as false. When these
packages are installed in silent mode, the security level will
Note: More information may be found in the agent log file using the
command "swjob -a log hp46144-0340 @ hp46144:/".
medium
#uemcli --getSecurityLevel
Security level = medium
3. Linux
a. Create a shell that will
Install the rpm package onto the system that will run
the command rpm -ivh <full path to SecureCLI
.rpm package>
Run the command /opt/emc/uemcli-
<VERSION>/bin/setlevel.sh low|medium|l|m to set
the security level.
b. Verify the security level setting by retrieving the error log that
is stored at /opt/emc/uemcli-VERSION/bin/setlevel.log