Dell Compellent OracleVM OVM Best Practices

Dell Compellent Storage Center
with Oracle OVM Virtual Machines

Best Practices, Tips and Tricks
Dell Compellent Storage Center with Oracle OVM Virtual Machines Best Practices
Document revision
Date
Revision
Comments
03/21/2012
Initial Draft
THIS BEST PRACTICES GUIDE IS FOR INFORMATIONAL PURPOSES ONLY, AND MAY CONTAIN
TYPOGRAPHICAL ERRORS AND TECHNICAL INACCURACIES. THE CONTENT IS PROVIDED AS IS, WITHOUT
EXPRESS OR IMPLIED WARRANTIES OF ANY KIND.
2012 Dell Inc. All rights reserved. Reproduction of this material in any manner whatsoever without
the express written permission of Dell Inc. is strictly forbidden. For more information, contact Dell.
Dell, the DELL logo, the DELL badge, and Compellent are trademarks of Dell Inc. Other trademarks and
trade names may be used in this document to refer to either the entities claiming the marks and names
or their products. Dell Inc. disclaims any proprietary interest in trademarks and trade names other than
its own.
Page 2
Contents
Document revision ................................................................................................. 2
Contents .................................................................................................................. 3
General syntax ..................................................................................................... 6
Conventions ......................................................................................................... 6
Overview ................................................................................................................. 7
Prerequisites ........................................................................................................ 7
Intended audience ................................................................................................. 7
Introduction ......................................................................................................... 7
Customer aupport.................................................................................................. 7
A note on Fibre Channel zoning ...................................................................................... 8
Host Bus Adapter settings ............................................................................................. 9
Qlogic Fibre Channel card BIOS or Module Settings .............................................................. 9
Emulex Fibre Channel card BIOS or Module Settings ............................................................. 9
Host Bus Adapter Queue Depth .................................................................................. 9
Putting it all together: ............................................................................................ 9
Connecting your VM servers to the Dell Compellent array: .............................................. 10
Dell Compellent Storage Center connections ............................................................... 10
Legacy ports ................................................................................................. 10
iSCSI.... ............................................................................................................ 11
Dell Compellent ports summary ............................................................................... 11
Overview of Dell Compellent Storage Center features ........................................................ 12
Data Instant Replay (DIR) ....................................................................................... 12
Data Progression ................................................................................................. 12
Dynamic Capacity (Thin Provisioning) ........................................................................ 12
Mapping Volumes to the Oracle VM Servers. ..................................................................... 13
Figure 4. Lun Selection on Dell Compellent ........................................................................ 14
Linux Multipathing .................................................................................................... 14
Multipathing on Dell Compellent................................................................................... 15
Boot from SAN ......................................................................................................... 16
Boot from SAN configuration ....................................................................................... 16
Introduction to Oracle VM ........................................................................................... 17
Oracle VM architecture .............................................................................................. 17
Establishing Oracle VM Infrastructure ............................................................................. 18
Page 3
Implementation tasks overview .................................................................................... 19
Fibre Channel topology .............................................................................................. 19
iSCSI topology ......................................................................................................... 20
Oracle VM 3.0.3 installation considerations ..................................................................... 20
Oracle OVM installation tips ........................................................................................ 21
Physical and Logical networks: ..................................................................................... 21
Installing the Oracle VM software ................................................................................. 22
A visual guide through the Installation....................................................................... 22
Oracle VM Server install ........................................................................................ 22
CD found ............................................................................................................... 23
Keyboard selection ................................................................................................... 24
OVS 3.0 Eula ........................................................................................................... 25
System to upgrade .................................................................................................... 26
Partitioning type ...................................................................................................... 27
Warning ................................................................................................................. 28
Review partition layout.............................................................................................. 29
Partitioning ............................................................................................................ 30
Boot Loader configuration .......................................................................................... 31
Arrange boot drives .................................................................................................. 32
Oracle VM Management Interface ................................................................................. 33
Oracle VM Management LAN ........................................................................................ 34
IPv4 configuration for eth0 ......................................................................................... 36
Miscellaneous network settings .................................................................................... 37
Hostname configuration ............................................................................................. 38
Timezone selection................................................................................................... 39
Oracle VM Agent password .......................................................................................... 40
Root password ......................................................................................................... 41
Installation to begin .................................................................................................. 42
Complete ............................................................................................................... 43
VM Server console .................................................................................................... 44
Login console .......................................................................................................... 45
Installing Oracle VM Manager ....................................................................................... 47
Steps to perform after Oracle VM Manager install .............................................................. 49
Testing a virtual machine migration .............................................................................. 55
iSCSI ..................................................................................................................... 56
Page 4
Live Volume overview................................................................................................ 58
Migrating the cluster with Dell/Compellent Live Volume ..................................................... 58
Live Volume failover ............................................................................................ 58
Using the Command Line with OVM to create a Multipath Live Volume ............................... 59
Index A... ......................................................................................................... 63
Glossary ................................................................................................................ 63
Index B.................................................................................................................. 68
Document references ........................................................................................... 68
Page 5
General syntax
Table 1.
Document syntax
Item
Convention
Menu items, dialog box titles, field names, keys
Bold
Mouse click required
Click:
User Input
Monospace Font
User typing required
Type:
Website addresses
http://www.compellent.com
Email addresses
info@compellent.com
Conventions
Notes are used to convey special information or instructions.
Timesavers are tips specifically designed to save time or reduce the number of steps.
Caution indicates the potential for risk including system or data damage.
Warning indicates that failure to follow directions could result in bodily harm.
Page 6
Overview
Prerequisites
This document assumes the reader has training or understanding of the following.
Operation and configuration of the Dell Compellent Storage Center
Linux operating systems
Oracle software installations and virtualization concepts
Intended Audience
This is a technical document for system administrators, and other information technology professionals
who are interested in learning or responsible for the setup and maintenance of Oracle OVM Enterprise
Linux Clusters and how they integrate with Dell Compellent storage.
Introduction
This document will provide configuration examples, tips, recommended settings and other storage
guidelines a user can follow while integrating Oracles OVM virtualization solution with the Storage
Center. The document has been written to answer many frequently asked questions with regard to
how OVM interacts with Dell Compellent Storage Centers various features, such as Dynamic
Capacity, Data Progression, and Data Instant Replays. OVM stands for Oracle Virtual Machines, and
is based on Xen Hypervisor technology. The document describes the infrastructure needed, as well
as the OVM installation process. It is recommended that you obtain the following documents from
Oracle and familiarize yourself with basic system components. Below are the URLs for OVM
Getting started guide, OVM Installation and upgrade, and OVM user guide.
http://docs.oracle.com/cd/E26996_01/e26529.pdf
http://docs.oracle.com/cd/E11081_01/doc/doc.21/e10899.pdf
Please note that the information contained within this document is intended only to be general
recommendations and may not be adaptable to all configurations. There are certain circumstances and
environments where the configuration may vary based upon individual or business needs.
Customer Support
Dell Compellent provides live support 1-866-EZSTORE (866.397.8673), 24 hours a day, 7 days a week,
365 days a year. For additional support, email Dell Compellent at support@compellent.com. Dell
Compellent responds to emails during normal business hours.
Page 7
A note on Fibre Channel zoning

Zoning Fibre Channel switches for OVM hosts is essentially no different than zoning any other hosts to
the Storage Center. Below are some points and options.
With Single Initiator Multiple Target Zoning, each Fiber Channel zone created should
have a single initiator (HBA port) and multiple targets (Storage Center front-end ports).
This means that each HBA port needs its own Fiber Channel zone containing itself and
the Storage Center front-end ports. Zoning OVM hosts by either port (commonly
referred to as hard zoning) or WWN (commonly referred to as soft zoning) is
acceptable.
With Port Zoning, If the environment has multiple fabrics, the additional HBA ports in
the host should have separate unique zones created in their respective fabrics.
With WWN Zoning, the zone only needs to contain the host HBA port and the Storage
Center front-end primary ports. In most cases, it is not necessary to include the
Storage Center front-end reserve ports because they are not used for volume
mappings.
Dell Compellent introduced Virtual Ports in Storage Center 5.0. Virtual Ports allow all
front-end IO ports to be virtualized. All front-end IO ports can be used at the same
time for load balancing as well as failover to another port. Virtual Ports are available
for Fibre Channel and iSCSCI connections only.
Page 8
Host Bus Adapter Settings

Qlogic Fibre Channel card BIOS or Module Settings
The connection options field should be set to 1 for point to point only
The login retry count field should be set to 30 attempts
The port down retry count field should be set to 5 attempts for faster failover
The link down timeout field should be set to 60 seconds.
The queue depth for the QLogic cards being controlled by the qla2xxx module, the
parameter that needs to be set is ql2xmaxqdepth. By default it is set to 32, but should
be changed to 128.
Emulex Fibre Channel card BIOS or Module Settings
The Node Time Out field lpfc_nodev_tmo field should be set to 60 seconds.
The topology field should be set to 2 for point to point only.
The queuedepth field should be set to 128
The following sections explain how the queue depth is set in each of the layers in the event it
needs to be changed.
The appropriate queue depth for a host may vary due to a number of factors, so it is
recommended to only increase or decrease the queue depth if necessary.
Host Bus Adapter Queue Depth

When configuring the host bus adapter for the first time, as mentioned previously, the queue
depth should be set to 128. This is because the driver module loaded for each HBA in the
system ultimately regulates the HBAs queue depth. For example, if the HBA BIOS is set to 255
and the driver module is set to 128, the maximum queue depth for that card or port will be
128.
Please see Dell Compellents Linux Best Practices document for a detailed explanation of how
to set these module parameters on your Oracle Enterprise Linux systems. The version number of
OEL corresponds to the version number of Redhat I.E. Oracle Linux 5 uses the same kernel as
Redhat Linux 5.0
Putting it all together:

The next few pages present a brief tutorial on some of Storage Centers features, as well how
to get connected to the Dell Compellent in order to start the installation process.
Page 9
Connecting your VM servers to the Dell Compellent array:

Dell Compellent Storage Center connections
After reviewing server-specific functions, you must understand various connection options
available with the Dell Compellent Storage Center that allow multiple paths to be presented to
the servers. The following section provides a quick overview and explains the differences
between Legacy Ports and Virtual Ports. This overview does not take the place of the Storage
Center Connectivity Guide available on http://knowledgecenter.compellent.com
Legacy Ports
In legacy mode, front-end IO ports are broken into primary and reserve ports based on a fault
domain. Primary/reserved ports allow IO to use the primary path; the reserve port is in a
standby mode until a primary port fails over to the reserve port. In terms of Linux Multipathd,
this requires twice the IO ports to enable multiple paths. Even more ports are required for a
dual fabric. Figure 1 illustrates dual-fabric failover ports in legacy mode.
Figure 2.
Legacy Primary/Reserved configuration
Figure 2 shows four fault domains to the CT-SC040. Each fault domain has a primary and
reserved port. For redundancy, a primary port connects to one controller; the reserved port in
that fault domain connects to the other controller. While this is a highly robust failover
solution, it requires a large number of ports. For this reason, Dell Compellent implemented
Virtual Ports.
Page 10
Virtual Ports
Dell Compellent introduced Virtual Ports in Storage Center 5.0. Virtual Ports allow all front-end
IO ports to be virtualized. All front-end IO ports can be used at the same time for load
balancing as well as failover to another port. Virtual Ports are available for Fibre Channel
connections only, iSCSI connections only, or both Fibre Channel and iSCSI.
Fibre Channel
To use Virtual Ports for Fibre Channel, all Fibre Channel switches and HBAs must support NPIV
(N_Port ID Virtualization).
Figure 3.
Virtual Ports configuration
Figure 3 shows a Virtual Port configuration. All ports of one transport type (Fibre Channel or iSCSI) are
in the same fault domain. All ports are active. If one port fails, the load is distributed between the
remaining ports. Failover is just as robust as shown in Figure 2, with half the ports.
iSCSI
iSCSI follows the same wiring and port setup as Fibre Channel with the exception of the Control
Port. iSCSI uses a Control Port configured for each of the Fault Domains. Servers connect to the
Control Port, which then redirects traffic to the appropriate Virtual Port. When configuring
Multipathd, this looks slightly different than with the Legacy Mode configuration because you
need to assign only the Control Port in the iSCSI Initiator software. These differences are
covered later in the OS-specific sections.
Dell Compellent ports summary

There are two methods to enable multiple paths from the Dell Compellent Storage Center,
Legacy Ports and Virtual Ports. Virtual Ports are preferred because they reduce the number of
HBAs required and thus reduce cost and overhead. Note that Virtual Ports are enabled based on
protocol: you can enable Virtual Ports on Fibre Channel only, iSCSI only, or both.
Page 11
Overview of Dell Compellent Storage Center features

Here are the definitions for some of the Dell Compellent storage arrays feature sets. Please
consult the best practices and tech tips documents to see how each of these features is
applied.
Data Instant Replay (DIR)

Data Instant Replay is most compared to "snapshot" technology. This is the ability to create
point-in-time copies where further changes to a volume of data are journaled in a way that
allows the volume to be rolled back to its original state when the point-in-time copy was taken.
These point-in-time copies can be mounted as volumes of data for the sake of partial data
restore as well as full volume restore. Of course, these point-in-time copies have various other
uses in an organization but for the sake of this paper we will focus on backup/recovery.
Data Instant Replay is one of many features that Dell Compellent offers. This feature gives you
the ability to make a point-in-time snapshot of your volumes (filesystems). There is no limit on
the number of Instant Replays taken.
Data Progression
Data Progression is another Dell Compellent Storage Center feature. This feature allows your
least frequently used data to migrate to lower tier (less-expensive SATA) disks, therefore
saving space on the higher tier (expensive FC/SAS) disks.
With the Data Progression feature enabled on the Dell Compellent system, you dont have to
worry about disk space taken up by data that has not been accessed. You can change the
number of days on a volume to alert Data Progression when to move the data. The default is 12
days. Vice-versa, if the data has been accessed for a certain amount of cycles, the data will
then migrate back to higher tier disks for performance. This also can be changed based on your
preference.
Dynamic Capacity (Thin Provisioning)

With traditional storage systems, administrators must purchase, allocate and manage capacity
upfront, speculating where to place storage resources and creating large, underutilized
volumes with long term growth built in. This practice leaves the majority of disk space
allocated yet unused, and only available to specific applications.
Dell Compellent Thin Provisioning, called Dynamic Capacity, eliminates the allocated but
unused capacity that is an unfortunate by-product of traditional storage allocation methods.
Dynamic Capacity delivers the highest storage utilization possible by eliminating allocated but
unused capacity. Dynamic Capacity completely separates storage allocation from utilization,
enabling users to create any size virtual volume upfront, yet only consume actual physical
capacity when data is written by the application.
Key points to consider:

If Fast Track is licensed, then Dynamic Capacity is enabled by default and will be utilized
behind the scenes. No manual configuration is required.
Page 12
Drives with higher RPM provide higher overall random-access throughput and shorter response
times than drives with lower RPM.
Because of better performance, SAS or Fibre Channel drives with 15K RPM are always
recommended for storing Oracle datafiles and online redo logs.
Serial ATA and lower cost Fibre Channel drives have slower rotational speed and therefore
recommended for Oracle archived redo logs and flashback recovery area with Data Progression.
Remote Instant Replay (Remote Replication)

Remote Instant Replay replicates continuous snapshots, called Replays between local remote
sites, ensuring business continuity at a fraction of the cost of other replication solutions. Using
Dell Compellents wizard-based setup, users are only six clicks away from a replication
solution. Remote Instant Replay supports traditional Fibre Channel replication or cost-effective
IP-based (iSCSI) replication, which brings disaster recovery within reach of every budget.
Mapping Volumes to the Oracle VM Servers

Volume Mapping Concepts
When sharing volumes between OVM hosts for such tasks as virtual server migration, it is
important that each volume is mapped to each OVM host using the same Logical Unit Number
(LUN).
For example:
There are three Oracle VM hosts named OVM1, OVM2, and OVM3.
A new volume is created named "LUN10-vm-repository."
This volume must be mapped to each of the OVM hosts as the same LUN:
Volume: "LUN10-vm-repository" Mapped to OVM1 -as- LUN 10
Volume: "LUN10-vm- repository " Mapped to OVM2 -as- LUN 10
Volume: "LUN10-vm- repository " Mapped to OVM3 -as- LUN 10
For Volume Mapping in Storage Center versions 4.x and earlier, each mapping must be created
separately, each time specifying the same LUN, for each individual OVM host.
For Volume Mappings in Storage Center 5.x and later the mapping process is greatly automated
by creating a server cluster object. This will allow the volume to be mapped to multiple OVM
hosts at the same time, automatically keeping the LUN numbering consistent for all the paths.
Also in Storage Center versions 5.x and higher, the system can auto select the LUN number, or
a preferred LUN number can be manually specified from the advanced settings screen in the
mapping wizard.
Page 13
Figure 4. LUN selection on Dell Compellent
Linux Multipathing
Although Dell Compellent Storage Centers provide redundancy and failover with multiple
controllers and RAID, servers still need a way to spread the IO load and handle internal failover
from one path to the next. This is where Linuxs multipathd plays an important role.
Multipathing is the ability of a server to communicate using multiple physical connections
between the host bus adapters in the server and the other device, typically in Fibre Channel,
or iSCSI SAN environments and direct attached storage when multiple channels are available.
Multipathing is managed at the device level. Multipath devices can be accessed by
/dev/mapper/ directory. Multipath in Linux requires installation of the device-mapper rpm
package. To check if the package is installed you can use the command rpm -qa | grep
multipath.
Because we are using a Dell Compellent storage subsystem that is automatically detected, no
further configuration of the multipath-tools is required. Otherwise you need to create
/etc/multipath.conf, and add an appropriate device entry for your storage subsystem. See
/usr/share/doc/packages/multipath-tools/multipath.conf.annotated for a template with
extensive comments.
Device mapping multipath IO features automatic configuration of the subsystem for a large
variety of setups. Active/passive or active/active (with round-robin load balancing)
configurations of up to eight paths to each device are supported.
Multipath-tools take care of automatic path discovery and grouping as well as automated path
retesting, so that a previously failed path is automatically reinstated when it becomes healthy
again. This minimizes the need for administrator attention in a production environment.
Multiple physical connections exist between host bus adapters in the server and the target.
Typical connection problems involve faulty adapters, cables, or controllers.
To determine which SCSI devices/paths correspond to the same LUN, the DM initiates a SCSI
inquiry. The inquiry response, among other things, carries the LUN serial number. Regardless of
the number paths a LUN is associated with, the serial number for the LUN will always be the
same. This is how multipathing SW determines which and how many paths are associated with
each LUN.
Querying the multipath IO status outputs the current status of the multipath maps.
The command is multipath -l.
Page 14
# multipath -l
360060160f390250024e60845be67df11 dm-2 DGC,RAID 5
[size=600G][features=1 queue_if_no_path][hwhandler=1 COMPELNT]
\_ round-robin 0 [prio=-2][active]
\_ 4:0:1:1 sdn
8:208 [active][undef]
\_ 3:0:1:1 sdf
\_ round-robin 0 [prio=-2][enabled]
\_ 4:0:0:1 sdj
\_ 3:0:0:1 sdb
Here you can see the output of the multipath l command, which shows the unique id and the
disk devices associated with the mpath device dm-2.
Please see Linux best practices for more multipath detail.
Multipathing on Dell Compellent

The first step in multipathing a volume is to create the necessary mappings. While a volume
can be configured as multipath with only one path, obviously in order to achieve the benefits,
it is necessary to have at least two paths. In this example, the server has two Fibre Channel
ports and the Dell Compellent has two front end ports on each controller. They are zoned in
two separate VSANs to simulate a dual switch fabric. After selecting the server to map the
volume too, the wizard will prompt for which ports on the server to map to.
The next screen selects the front end ports on the Dell Compellent to use for the mapping.
Select both ports from one controller.
Page 15
Lastly assign the LUN. In this case the Dell Compellent presents four pairs of mappings. In
reality there are only two valid paths, however the Dell Compellent system cannot determine
which are valid or not. It its best to create all the mappings unless the correct WWN pairings
are known. Now with the Volumes assigned to the OVM hosts, Lets look at SAN boot.
Boot from SAN

There is an ongoing discussion about whether or not to boot hosts from SAN. In some cases,
such as with blade servers that do not have internal disk drives, booting from SAN may be the
only option, but a lot of OVM hosts can have internal mirrored drives giving the flexibility to
choose. The benefits of booting from SAN are obvious. It alleviates the need for internal drives
and allows the ability to take replays of the boot volume. Also if the Oracle VM Servers were
boot from SAN, it would allow for migration of the entire cluster architecture via Dell
Compellents Live Volume feature. This ability can also be nice for server systems
maintenance.
Since the decision to boot from SAN depends on many business-related factors including cost,
recoverability, and configuration needs, we have no specific recommendation. The Live
Volume section of this document assumes the hosts are booting from the San.
Boot from SAN Configuration

When deciding to boot OVM hosts from SAN, there are a few best practices that need
consideration. In Storage Center 4.x and earlier, when initially mapping the boot volume to the
OVM host, the mapping wizard will allow the selection of the individual paths, one must be
sure to specify LUN 0.
Page 16
Once the boot Volumes are added to the Oracle VM systems and have the systems are
configured to boot from them, the Oracle VM Server installation process can begin.
Introduction to Oracle VM
Software-based server virtualization solutions for X86

Server partitions and consolidation
High availability and scalability
Fast deployment and standardization
Provide the virtualization support for grid/cloud
Oracle VM Architecture
Oracle VM Server is Xen Hypervisor.

Management Linux Kernel.
Support of devices, IO, Networkingetc.
VM: Guest OS with applications running on DomU.
Virtual Server Pool: An autonomous region of VM Servers.
Oracle VM Manger: Gui management solution for Oracle VM.
Oracle VM Agent: Installed with VM server and communicating with Oracle VM Manager.
Dom0: An abbreviation for domain zero. The management domain with privileged
access to the hardware and device drivers.
DomU - An unprivileged domain where VMs live with no direct access to the hardware
or device drivers.
Page 17
Establishing Oracle VM Infrastructure

Virtual System Architecture Design:
Page 18
One Dell x86 server for installation of OVM Server Manager software
Two or more Dell x86 servers with matching CPU speed, Stepping, and RAM for
installation of Oracle VM server software (hypervisors).
Two or more Virtual Machines running Oracle Enterprise Linux.
Dell Compellent Storage Array running version 5.4.4 of the Dell Compellent OS, and
configured with Virtual Ports.
Implementation Tasks Overview
VM Server installation and VM Manager installation.

Connect VM Servers to the VM Manager.
Virtual Infrastructure network and storage configuration.
Create guest VMs using ISO image.
Configure Network and Shared Storage on VMs.
Fibre Channel Topology

Example of Virtual Infrastructure based on Oracle VM 3.0 (FC storage)
In this illustration we see the VM Management servers using two HBA ports for redundant paths,
two network interfaces being used for private VM networks, which can also communicate
heartbeat signals. One public addressable network for administering the VM Servers, and a
management network for doing Virtual Machine migrations. Bond0 is created by the VM Server
Page 19
install and cannot be changed. The VMs also have NIC card plummed to use the Bond2 device
over the private net, as well as a device to reside on the public net so the VMs are available
externally to the frame.
iSCSI Topology
iSCSI is used to allocate volumes to the virtual machines so they appear on the VM host as SCSI
disks. When allocating disk volumes to the virtual machines through the OVM Hypervisors, they
appear as IDE disks, which in this case is not desirable for things like DM-Multipathetc. The
Dell Compellent array works really well for this task, as it is possible to present volumes to the
cluster with Fiber Channel and later present Virtual Machines with volumes using iSCSI.
Oracle VM 3.0.3 installation considerations

Physical Machines: For a minimal setup three servers will be needed, one for Oracle Virtual
Manager and two for Oracle VM servers (Hypervisor Platforms).
It is vitally important that the Oracle VM server machines be the same model server with the
same CPU Stepping and the same amount of RAM. If the Stepping is different it will not be
possible to perform any type of Virtual Machine live migrations!
Virtual Machines: A minimum of two Virtual Machines are needed to build an OVM cluster
infrastructure. (one VM per OVM server works nicely). It is possible to Clone the VMs at
different points during progress to preserve all the VMs files and settings. Creating continual
Replays of the Cluster Volumes on the Dell Compellent will help to recover the systems should
anything get corrupted.
Page 20
Oracle OVM Installation Tips

Please download and read Oracle VM Getting Started Guide and for detailed instructions on
the steps listed below, also take a look at Oracle VM Installation and Upgrade Guide. The
actual order of how things are done is important as well, and this document addresses these steps
in the proper order.
1.) Install the two OVM hypervisor servers using the Oracle Virtual Server 3.0.3 installation boot
media.
2.) Install the third server with the Oracle Virtual Manager 3.0.3 installation boot media.
3.) Make sure to record IP addresses you will be using for the NICs on all three systems, and that
all three primary NICs are on the same subnet segment. It is a good idea to keep a journal of all
the information for each of the servers physical and virtual IP addresses and other important
information.
Physical and Logical Networks:
OVM servers should have at least 4 NIC ports, one for primary and three inter connects.
One NIC port for Primary Network to administer and manage the servers.
One NIC port for Heartbeat inter connect to the secondary OVM server.
One NIC port for Virtual Machines Private Network traffic.
One NIC port for Virtual Machine Migrations traffic.
An example of Network and Fibre connections in an OVM Cluster.

Page 21
Installing the Oracle VM software

A visual guide through the Installation
After creating a SAN volume to install OVM server software on, and making sure you have all
the proper hardware installed (Fiber Channel HBAs etc.) boot the server to the install media.
Oracle VM Server Install

After booting to the Oracle VM Server Installation media the Oracle VM Server screen is
displayed. Press Enter to begin the installation.
The installer is only available in text mode.
Page 22
CD Found
The CD Found screen is displayed. If you want to make sure the CD was created correctly you
can have the installer test it for errors.
To test the CD, select OK and press Enter. The CD is tested and any errors are reported. To
skip media testing and continue with the install, select Skip and press Enter.
Page 23
Keyboard Selection
Next the Keyboard Selection screen will be displayed. Select the keyboard layout (for
example,, us for U.S. English) from the list of available options. The keyboard you select
becomes the default layout for the operating system.
Select OK and press Enter.
Page 24
OVS 3.0 Eula

The Oracle VM Server 3.0 End User License Agreement screen is displayed. Read the license
agreement text and click Accept to accept the terms of the agreement and continue the
installation.
Page 25
System to Upgrade
If an existing Oracle VM Server is found on the computer, the System to upgrade screen is
displayed. Select Oracle VM Server 3.x (disk) to upgrade the existing installation. Upgrade
instructions are described in section 4.3.2 of Upgrading Oracle VM Server Using the CD.
Upgrading an Oracle VM Server from release 2.x is not supported; you must perform a fresh
install.
Page 26
Partitioning Type
Next the Partitioning Type screen is displayed. Select whether you want to:
Remove all partitions and create a default Partition layout.

Remove all Linux Partitions and Create a new default Partition Layout.
Use the free space on selected drives to create a new default layout.
Create a minimal Partition laypout for installation to a USB drive.
Create a custom partition layout.
Oracle recommends to use a default partition layout.

As the Oracle VM installer can only be run in text mode, you cannot setup LVM (Logical Volume
Manager). If you want to create an LVM configuration, press Alt+F2 to use the terminal and
run the lvm command. To return to the Oracle VM Installer, press Alt+F1.
If you selected to remove a partition, a Warning screen is displayed to confirm that you want to
remove the partition(s), including the data contained on any partitions.
Page 27
Warning
Just a little warning that you are about to initialize this drive, and that any current data will
be removed.
If this is ok then Select Yes and press Enter.
Page 28
Review Partition Layout

The Review Partition Layout screen is displayed. If you do not want to review the partition
layout, select No and press Enter.
If you want to review the partition layout, select Yes and press Enter.
Page 29
Partitioning
The Review Partition Layout screen is displayed. Review the partition layout and make any
changes needed.
Select OK and press Enter to save any changes.
Page 30
Boot Loader Configuration

The Boot Loader Configuration Screen is displayed. If necessary change the drive for the
master boot record by clicking the Change Drive Order button and selecting a different drive.
Page 31
Arrange Boot Drives

Select a drive in the list and click the Move up and Move down buttons to change the order,
making sure that the preferred drive to have the Master Boot Record is at the top. Select Back
and press Enter to return to the Boot Loader Configuration screen.
Page 32
Oracle VM Management Interface

The Oracle VM Management Interface screen is displayed. Select the network interface to use
for management of the computer. Then choose one of these options:
Select OK to connect the interface to a regular network.

Select Add to VLAN to connect the interface to a VLAN-type network.
Press Enter.
For non-VLAN installation, skip next page and move forward to the IPv4 Configuration for eth0
screen.
Page 33
Oracle VM Management LAN

If you selected Add to VLAN, the Oracle Management VLAN screen is displayed. If your Oracle
VM Management network is on a VLAN, enter the applicable VLAN tag. Note that 0 is reserved
and that 1 corresponds with untagged.
Page 34
Oracle VM Cluster network diagram with a high level view showing VLANs and a VLAN group,
and how the switch might look with trunk ports.
You can skip this function and continue if you do not wish to configure VLANS at this time.
Page 35
IPv4 Configuration for eth0

Next the IPv4 Configuration for eth0 screen is displayed. If your computer uses a static IP
address, choose manual address configuration to enter the IP address and prefix (Netmask) for
your computer. If your computer uses DHCP to assign its IP address, select Dynamic IP
configuration (DHCP).
Page 36
Miscellaneous Network Settings

The Miscellaneous Network Settings screen is displayed. Enter the gateway, Primary DNS and
optional Secondary DNS in the respective fields.
Page 37
Hostname Configuration
The Hostname Configuration screen is displayed. If your machine has its own hostname, select
manually and enter the hostname or IP address. Please use a fully qualified hostname, for
example, myserver.example.com. If your machine uses DHCP to assign its hostname, select
automatically via DHCP.
Page 38
Time Zone Selection

The Time Zone Selection screen is displayed. If your operating system uses Coordinated
Universal Time (UTC), select System clock uses UTC. Select your time zone by selecting the
city closest to your computers physical location.
Page 39
Oracle VM Agent Password

The Oracle VM Agent Password screen is displayed. Enter a password to be used for the Oracle
VM Agent in the Password field. The password is used by Oracle VM Manager to manage and
monitor Oracle VM, and the guests created and running within it. This is the password to use
when discovering Oracle VM Servers in Oracle VM Manager. Re-enter the password in the
Password (confirm) field.
The password characters are not echoed to the screen.
Select OK and press Enter. If the two passwords do not match, the installer prompts you to
enter them again.
If you need to change this password after install, use the ovs-agent-passwd utility.
Page 40
Root Password
The Root Password screen is displayed. Enter a password for the root user in the Password
field. The root password must be at least six characters long. Re-enter the password in the
Password (confirm) field. The password characters are not echoed to the screen.
Page 41
Installation to Begin
The Installation to Begin screen is displayed. The installer installs and configures Oracle VM.
Once the Enter key is pressed here, the installer will format all the partitions and install the
VM Server Image onto the drive selected. The VM Server installation only takes ten to fifteen
minutes depending on the servers hardware.
Page 42
Complete
When all the files are installed and the configuration is complete, the Complete screen is
displayed. Please remove the Oracle VM CD from the optical drive in the machine at this time.
Select Reboot and press Enter to restart the VM Server.
Page 43
VM Server Console
After reboot, the Oracle VM status console is displayed. The status console displays useful
information about the servers hardware and configuration as well as its cluster and server pool
membership. Notice that there will not be much information in these fields as of yet. Once the
Oracle Manager server is installed, and the cluster is configured, these fields will all show all
the current VM Server information and status of the cluster.
Page 44
Login Console
To log on to the Oracle VM, press ALT+F2 to display the login console. You can login to the
Oracle VM as root, with the password you set during the install. To switch back to the status
console, press Alt+F1.
The Oracle VM installation is complete. The Oracle VM Agent is started automatically and
restarts each time the computer is rebooted.
A log of the install is located in /root/install.log. If the install fails, review this log file to
help identify the cause.
Page 45
This is another view of the overall architecture and anatomy of an Oracle VM cluster. The Web GUI
seen at the top of the diagram is the Web User Interface (UI), which resides on the Oracle VM
Manager server.
Page 46
Installing Oracle VM Manager

Oracles VM Manager has a text-based installer that asks a series of questions, and then based
on the answers performs the scripted install. Once the Installation media is mounted on the
server, CD to the directory where the disk is mounted and run the runInstaller.sh script. The
example below is for doing a Demo install. The production install is essentially the same, but
includes a few more questions about the Oracle install to be performed after the installation of
Oracle VM Server software. Output from server is in blue.
# ./runInstaller.sh
A set of installer options are displayed:
Please select an installation type:
1: Demo
2: Production
3: Uninstall
4: Help
At the command prompt, enter 1 (a demo install) or 2 (a production install) to install Oracle
VM Manager. If you already have Oracle VM Manager installed and want to uninstall it, select 3
(uninstall).
Starting demo installation ...
The Demo installation type will use an XE database.
The usage of XE is for *demo purposes only* and is not supported for production.
Please *do not* plan to start with XE and migrate to a supported version of the database as this
may not be possible.
For production environments or any long term usage please use the "Production" option with an SE
or EE database.
1: Continue
2: Abort
Select Number (1-2): 1
You have the option to continue or abort the installation. Enter 1 to continue.
Enter a password to use for all users created during the demo installation.
Verifying installation prerequisites ...
*** WARNING: Recommended memory for the Oracle VM Manager server installation is 3950 MB RAM
One password is used for all users created and used during the installation.
Enter a password for all logins used during the installation:
Enter a password for all logins used during the installation (confirm):
Verifying configuration ...
Warnings may occur if not all recommendations are met, as you can see in the screen
output above. If one of the prerequisites is not met, the installation will abort.
You have the option to continue or abort the installation. Enter 1 to continue. Progress
of the installation is output on your screen step by step, as shown below:
Start installing the configured components:
1: Continue
Page 47
2: Abort
Select Number (1-2): 1
Step 1 of 9 : Database ...
Installing Database ...
Retrieving Oracle Database 11g XE ...
Installing Oracle Database 11g XE ...
Configuring Oracle Database 11g XE ...
Step 2 of 9 : Java ...
Installing Java ...
Step 3 of 9 : Database Schema ...
Creating database schema 'ovs' ...
Step 4 of 9 : WebLogic ...
Retrieving Oracle WebLogic Server 11g ...
Installing Oracle WebLogic Server 11g ...
Step 5 of 9 : ADF ...
Retrieving Oracle Application Development Framework (ADF) ...
Unzipping Oracle ADF ...
Installing Oracle ADF ...
Step 6 of 9 : Oracle VM ...
Retrieving Oracle VM Manager Application ...
Extracting Oracle VM Manager Application ...
Installing Oracle VM Manager Core ...
Step 7 of 9 : Domain creation ...
Creating Oracle WebLogic Server domain ...
Starting Oracle WebLogic Server 11g ...
Configuring data source 'OVMDS' ...
Creating Oracle VM Manager user 'admin' ...
Step 8 of 9 : Deploy ...
Deploying Oracle VM Manager Core container ...
Deploying Oracle VM Manager UI Console ...
Deploying Oracle VM Manager Help ...
Enabling HTTPS ...
Granting ovm-admin role to user 'admin' ...
Step 9 of 9 : Oracle VM Manager Shell ...
Retrieving Oracle VM Manager Shell & API ...
Extracting Oracle VM Manager Shell & API ...
Installing Oracle VM Manager Shell & API ...
Retrieving Oracle VM Manager Upgrade tool ...
Extracting Oracle VM Manager Upgrade tool ...
Installing Oracle VM Manager Upgrade tool ...
Copying Oracle VM Manager shell to '/usr/bin/ovm_shell.sh' ...
Installing ovm_admin.sh in '/u01/app/oracle/ovm-manager-3/bin' ...
Installing ovm_upgrade.sh in '/u01/app/oracle/ovm-manager-3/bin' ...
Enabling Oracle VM Manager service ...
Shutting down Oracle VM Manager instance ...
Restarting Oracle VM Manager instance ...
Waiting 15 seconds for the application to initialize ...
Oracle VM Manager is running ...
Oracle VM Manager installed.
Page 48
When the installation is complete a summary similar to the following is displayed:
Oracle VM Manager installed.
Installation Summary
-------------------Database configuration:
Database host name : localhost
Database instance name (SID): XE
Database listener port : 1521
Application Express port : 8080
Oracle VM Manager schema : ovs
Weblogic Server configuration:
Administration username : weblogic
Oracle VM Manager configuration:
Username : admin
Core management port : 54321
UUID : 0004fb0000010000cc1b57e39ff77ea9
Passwords:
There are no default passwords for any users. The passwords to use for Oracle VM
Manager, Oracle Database 11g XE, and Oracle WebLogic Server have been set by you
during this installation. In the case of a demo install, all passwords are the
same.
Oracle VM Manager UI:
http://myserver.example.com:7001/ovm/console
https://myserver.example.com:7002/ovm/console
Log in with the user 'admin', and the password you set during the installation.
Please note that you need to install tightvnc-java on this computer to access a virtual
machine's console.
For more information about Oracle Virtualization, please visit:
http://www.oracle.com/virtualization/
Oracle VM Manager installation complete.
Steps to perform after Oracle VM Manager install

Login to http://HOSTNAME:7001/ovm/console/faces/login.jspx where HOSTNAME is the network
addressable name of the server that the Oracle Virtual Manager software was installed on. The
username and password for login were created during the install. This is a Web GUI for creating and
administering server pools, repositories, virtual machines, networks, storage, and RAM resources in the
Oracle VM cluster.
Page 49
The Oracle VM Manger web GUI Interface.

Details for the following steps are in the Oracle VM Installation and Upgrade Guide document
however, this is the proper order to execute them in the Oracle VM manager Web GUI.
1.)
2.)
3.)
4.)
5.)
6.)
7.)
8.)
9.)
Discover the two OVM Servers with the Discovery Tool.

Register the Dell Compellent Storage on the Storage tab in the Management GUI.
Create pool of 20 virtual NICs with the virtual NIC Address Creation Tool.
Add Primary Network and create Internal Network for Virtual Machines (This is an inter-connect
network between the two VM server hosts for Virtual Machine Migrations).
Create and name a Server Pool (bundles a Dell Compellent volume with a Virtual IP Address).
Create and name a Storage Repository with large Dell Compellent Repository Volume.
Add objects (ISOs and Templates) to the new Repository (Import ISOs via FTP server).
Create Virtual Machines (Dont forget to include enough disk space for Oracle install).
Install 64 bit Oracle Enterprise Linux version 5 Update 5 on the VMs.
Page 50
An example view of the web GUIs menu system, below is the default home view after logging in.
From the home view you can see the Server Pools and their Virtual Machines, and by clicking on any
Virtual Machine you can see its network, disk, repository and memory properties.
Page 51
In the jobs view you can see logs of every action that the administrator has taken since the system was
brought online.
Navigating in the home view Server Pools tab to a virtual machine.

Page 52
Above is the web admin GUI page for registering and configuring the Dell Compellent Storage Array.
The admin can create any name label and description, pick the storage type and plugin, and define the
controller port information.
In the Storage tab in the hardware view, the user can drill down to see the Cluster Volume groups and
Physical Disks (Volumes), as well as some properties of the Dell Compellent array being utilized.
Page 53
This is the Hardware View. It allows the user to monitor all the hardware in the cluster, view or create
new networks, view and create VLAN groups and watch event logs.
Page 54
Above on the Networks tab of the home view, the admin can use physical ports on the VM Servers to
create external or internal networks for maintenance, VM Migrations, or heartbeat Networks.
Testing a Virtual Machine Migration

Once the Virtual Machines are installed on each of the OVM servers, make sure to Migrate a Virtual
Machine back and forth between the two VM servers. This will not be possible if your CPUs do not have
the same Stepping. Stepping information will also be displayed in the server information details inside
the Management GUI. If Virtual Machine Migration goes well, then it is time to prepare the virtual
machines for the Oracle RAC installs.
Manual migration of virtual machines, bringing up servers on another host.
Page 55
Testing Live Migrations of Oracle Enterprise Linux Virtual Machines from one VM Server to another.
iSCSI
This is just a quick sample of how to get the iSCSI ports setup. For a much more detailed doc please
refer to Dell/Compellents Linux Best Practices documentation. A quick study of the iscsiadm man
page on the Linux host is also recommended.
1.) Make sure that the VM hosts network ports have been configured and can communicate with
your Dell Compellents iSCSI ports.
2.) The Oracle Enterprise Linux VMs need to have the the iscsi-initiator-utils RPM installed.
3.) Have the iSCSI daemon service start up at boot through the use of the /etc/init.d/iscsi init
script.
4.) Have Virtual Ports enabled on the Dell Compellent array.
5.) Run command iscsiadm -m discovery -t sendtargets -p <Compellents ISCSI controller IP>.
The main purpose of the discovery is to have the hosts iSCSI daemon save the iSCSI node
information in its /var/lib/iscsi file - so it will know how to login to the subsystem to bring all
your iSCSI volumes online at boot time.
6.) To login now without a reboot, run command iscsiadm m node login.
7.) On the Dell Compellent, create the iSCSI server Cluster Object and assign some Volumes to it.
8.) Scan the hosts for new Volumes with iscsiadm m session R This scan will make the newly
added Volumes available for use on the cluster host.
Page 56
Oracle VM Networking example of a large cluster utilizing many Server Pools.

This shows an example of an Oracle VM environment with split Network functions, and how a cluster
can be spread to many nodes. Each Oracle VM Server is connected to the management network,
regardless of which server pool they belong to.
Page 57
Live Volume overview

Live Volume is a new software option for Dell Compellent Storage Center that builds upon the Dell
Fluid Data architecture. Live Volume enables non-disruptive data access and migration of data between
two Storage Centers. Live Volume is a software-based solution integrated into the Dell Compellent
Storage Center Controllers. Live Volume is designed to operate in a production environment, allowing
both Storage Centers to remain operational during volume migrations.
Live Volume increases operational efficiency, reduces planned outages, and enables a site to avoid
disruption during impending severe weather. Live Volume provides these powerful new options:
Storage follows the application in virtualized server environments.
Live Volume automatically migrates data as virtual applications are moved.
Zero downtime maintenance for planned outages.
Live Volume enables all data to be moved non-disruptively between Storage Centers,
enabling full planned site shutdown without downtime.
On-demand load balancing. Live Volume enables data to be relocated as desired to
distribute workload between Storage Centers.
Live Volume is designed to fit into existing physical and virtual environments without disruption, extra
hardware requirements or any changes to configurations or workflow. Physical and virtual servers see a
consistent, unchanging virtual volume. All volume mapping is consistent and transparent before,
during, and after migration. Live Volume can be run automatically or manually and is fully integrated
into the Storage Center software environment. Live Volume operates asynchronously and is designed
for planned migrations where both Storage Centers are simultaneously available. A Live Volume can be
created between two Dell Compellent Storage Centers residing in the same Data center or between
two well-connected data centers.
Migrating the cluster with Dell Compellent Live Volume

One of the goals of this paper is to show how to create an OVM Infrastructure on Dell Compellent
arrays with Linux Virtual Machines, for the purpose of running a multi node RAC Database Instance.
The idea is that the whole cluster (not just the Virtual Machines) has the capability to fail over to
another physical Dell Compellent array with no Database outage - via Dell Compellents Live Volume
technology. In order for Live Volume to work, the paths to the volumes must be zoned to two or more
Dell Compellent arrays. Below is an illustration of basic Live Volume replication.
Live Volume failover

There are many documents illustrating and detailing the setup of Live Volume. Please see
Dell/Compellents Live Volume best practices and Storage Center migrations with Linux and Live
Volume for an in depth explanation of getting this set up. We will describe what needs to be done
on the OVM hosts from a multipath perspective on the Fibre Channel Volumes the cluster was created
with. The same logic will apply to any iSCSI volumes given to the Virtual Machines if they exist.
Page 58
Using the Command Line with OVM to create a Multipath Live Volume
To begin, we have an OVM server with a boot partition using Fiber Channel connectivity to its
Storage Center (I.E Boot from SAN). In this case, we will be moving the boot volume mpatha:
Then just follow up after doing the same to all the other cluster Volumes seen on the host. This
also assumes you have the host mapped to two different Storage Centers.
# multipath -ll
mpatha (36000d310000069000000000000000d36) dm-0 COMPELNT,Compellent Vol
size=16G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
|- 0:0:7:0 sda 8:0 active ready running
`- 1:0:4:0 sdb 8:16 active ready running
# lsscsi
[0:0:7:0] disk COMPELNT Compellent Vol 0504 /dev/sda
[1:0:4:0] disk COMPELNT Compellent Vol 0504 /dev/sdb
At this point the volume that makes up mpatha can be converted into a Live Volume. Once
that is done, the Linux server needs to rescan its HBAs in order to be visible during the server
object creation process on the destination Storage Center. Note that this means the server
must be able to reach both Storage Centers through the Fiber Channel fabric. To rescan the
HBAs:
Page 59
# echo "- - -" > /sys/class/scsi_host/host0/scan

The WWN of the HBAs should now be visible within the Storage Center interface. To cross
reference this, the Linux server can print the WWNs visible on the HBAs:
# cat /sys/class/scsi_host/host0/device/fc_host/host0/port_name
0x2100001b320fa28b
# cat /sys/class/scsi_host/host1/device/fc_host/host1/port_name
0x2101001b322fa28b
Once the server object is created, it can be mapped to the Live Volume. This mapping needs to
use the Advanced mapping option of mapping it as Lun0, as this is the boot volume for the
server, and will continue to be Lun0 on the new Storage Center. Once the mapping has been
created, the HBAs need to be rescanned for the new paths to appear:
The addition of the new paths can be confirmed with the multipath command:
# multipath -ll
|- 0:0:7:0 sda 8:0 active ready running
|- 1:0:4:0 sdb 8:16 active ready running
|- 0:0:1:0 sdc 8:32 active ready running
`- 1:0:2:0 sdd 8:48 active ready running
At this point, the server will be directing IO down each path, which means that both the
original and the destination Storage Center will be processing that servers I/O. The next step is
gracefully failing the paths to the original Storage Center with the multipathd command:
# multipathd fail path sda
ok
Confirm that the path was committed to a failed state:
# multipath -ll
|- 0:0:7:0 sda 8:0 failed ready running
|- 1:0:4:0 sdb 8:16 active ready running
Then fail the remaining path to the original Storage Center:
# multipathd fail path sdb
ok
And confirm the failure of that path:
# multipath -ll
Page 60
|- 0:0:7:0 sda 8:0 failed ready running
|- 1:0:4:0 sdb 8:16 failed ready running
At this point, all IO will be sent to the destination Storage Center. Since the path original paths
have been gracefully failed with no event, the paths can be deleted from the running multipath
configuration:
# multipathd remove path sda
ok
Confirm the removal of the path:
# multipath -ll
|- 1:0:4:0 sdb 8:16 failed ready running
Remove the final path to the original Storage Center:
# multipathd remove path sdb
ok
Confirm the removal of the final path:
# multipath -ll
The current running configuration of the multipath device mapper is such that the paths to the
original Storage Center will never be used. Now the drive objects that make up the original
paths can be gracefully removed:
# echo 1 > /sys/block/sda/device/delete
# echo 1 > /sys/block/sdb/device/delete
Because IO is going only to the destination Storage Center, the Live Volume logic will (if it has
not already) work to identify the destination Storage Center as the primary recipient of IO, and
will begin the process of making it the primary Storage Center for the Live Volume. Once the
destination Storage Center has taken the primary responsibilities for the Live Volume, the
server to volume mapping on the original Storage Center can be removed, and the Live Volume
converted back to a regular volume. Once the mappings have been removed, the HBAs should
be rescanned one final time:
The server can then be removed from the original Storage Centers fabric zone.
Page 61
Once the OVM server Volumes and the iSCSI Volumes are all converted to Live Volumes, the
complete cluster plus VMs can fail from one Dell Compellent Storage Center to another in
minutes.
Using Dell Compellent Enterprise Manager, a live Volume can be created from a new Volume,
an existing Volume, or an existing Replication.
Page 62
Live Volume replication attributes in Dell Compellents Enterprise Manager.
For complete and detailed information on how to get Live Volume configured, please see the Dell
Compellent Live Volume Best Practices document.
Index A
Glossary
A
Affinity - Specify that specific virtual machines should always run on the same host.
Page 63
Anti-affinity - Specify that specific virtual machines should never run on the same host.
Assembly - An infrastructure template containing a configuration of multiple virtual machines with
their virtual disks, and the inter-connectivity between them. Assemblies can be created as a set of
.ovf (Open Virtualization Format) and .img (disk image) files, or may all be contained in a single .ova
(Open Virtualization Format Archive) file.
D
Dell Compellent Data Instant Replay - Data Instant Replay is most compared to "snapshot" technology.
This is the ability to create point-in-time copies where further changes to a volume of data are
journaled in a way that allows the volume to be rolled back to its original state when the point-in-time
copy was taken.
Dell Compellent Data Progression - A Dell Compellent Storage Center feature. This feature allows your
least frequently used data to migrate to lower tier (less-expensive SATA) disks, therefore saving space
on the higher tier (expensive FC/SAS) disks.
Dell Compellent Dynamic Capacity (Thin Provisioning) With traditional storage systems, administrators
must purchase, allocate and manage capacity upfront, speculating where to place storage resources
and creating large, underutilized volumes with long term growth built in. This practice leaves the
majority of disk space allocated yet unused, and only available to specific applications.
Dell Compellent iSCSI Dells iSCSI follows the same wiring and port setup as Fibre Channel with the
exception of the Control Port. iSCSI uses a Control Port configured for each of the Fault Domains.
Servers connect to the Control Port, which then redirects traffic to the appropriate Virtual Port.
Dell Compellent Legacy Port Mode - In legacy mode, front-end IO ports are broken into primary and
reserve ports based on a fault domain. Primary/reserved ports allow IO to use the primary path; the
reserve port is in a standby mode until a primary port fails over to the reserve port.
Dell Compellent Virtual Port Mode - Allows all front-end IO ports to be virtualized. All front-end IO
ports can be used at the same time for load balancing as well as failover to another port. Virtual Ports
are available for Fibre Channel and iSCSCI connections only.
Dell Compellent Live Volume - A way to move or migrate Dell Compellent volumes between storage
Center systems without interuption to service..
Dell Compellent Volume Mapping The process of making a Dell Compellent Volume on the array
available to a specific server object(s).
Domain - A configurable set of resources, including memory, virtual CPUs, network devices and disk
devices, in which virtual machines run. A domain is granted virtual resources and can be started,
stopped and rebooted independently.
See also dom0 .
See also domU.
dom0 - An abbreviation for domain zero. The management domain with privileged access to the
hardware and device drivers. Dom0 is the first domain started by the Oracle VM Server at boot time.
Page 64
Dom0 has more privileges than domU. It can access the hardware directly and can manage the device
drivers for other domains. It can also start new domains.
domU - An unprivileged domain with no direct access to the hardware or device drivers. Each domU is
started by Oracle VM Server in dom0.
G
Guest - A guest operating system that runs within a domain in Oracle VM Server. A guest may be
paravirtualized or hardware virtualized. Multiple guests can run on the same Oracle VM Server.
H
Hardware virtualized machine See Virtual Machine.
Host computer a computer connected to a network.
Hypervisor - a hypervisor, also called virtual machine manager (VMM), is one of many hardware
virtualization techniques allowing multiple operating systems, termed guests, to run concurrently on a
host computer. It is so named because it is conceptually one level higher than a supervisory program.
The hypervisor presents to the guest operating systems a virtual operating platform and manages the
execution of the guest operating systems. Multiple instances of a variety of operating systems may
share the virtualized hardware resources. Hypervisors are very commonly installed on server hardware,
with the function of running guest operating systems, that themselves act as servers.
M
Management domain (a.) A virtual machine with an unmodified guest operating system. It is not
recompiled for the virtual environment. There may be substantial performance penalties running as a
hardware virtualized guest. Enables Microsoft Windows operating system to be run, as well as legacy
operating systems. Hardware virtualization is only available on Intel VT or AMD SVM CPUs.
(b.) The physical computer on which Oracle VM Server is installed.
(c.) The hypervisor, monitor, or Virtual Machine Manager (VMM). It is the only fully privileged entity in
the system. It controls only the most basic resources of the system, including CPU and memory usage,
privilege checks, and hardware interrupts.
See also dom0 .
Master Oracle VM Server - A component of Oracle VM Agent. An application that acts as the contact
point to Oracle VM Manager, and to other Oracle VM Agents. Provides virtual machine host loadbalancing, and local persistence for Oracle VM Server. There is only one master Oracle VM Server in a
server pool. A physical server can perform as the master Oracle VM Server, Utility Server and Virtual
Machine Server simultaneously.
N
Non-Sparse Copy - A clone of the type "non-sparse copy" is a disk image file of a physical disk, taking
up the space equivalent to the full specified disk size, including empty blocks.
See also Sparse Copy.
Page 65
O
Oracle VM Agent - An application installed with Oracle VM Server. It communicates with Oracle VM
Manager for management of virtual machines. Oracle VM Manager manages the virtual machines
running on Oracle VM Server by communicating with Oracle VM Agent. It contains three components:
master Oracle VM Server, Utility Server, and Virtual Machine Server.
Oracle VM Server - A self-contained virtualization environment designed to provide a lightweight,
secure, server-based platform for running virtual machines. Oracle VM Server is based upon an updated
version of the Xen hypervisor technology. Includes Oracle VM Agent to enable communication with
Oracle VM Manager.
Oracle VM Manager - Provides the user interface, which is an Application Development Framework
(ADF) web application, to manage Oracle VM Server pools. Manages virtual machine life cycle, including
creating virtual machines from templates or from installation media, deleting, powering off, uploading,
deployment and live migration of virtual machines. Manages resources including ISO files, templates
and shared virtual disks.
P
Paravirtualized machine - A Virtual Machine with a kernel that is recompiled to be made aware of the
virtual environment. Runs at near native speed, with memory, disk and network access optimized for
maximum performance.
Port Zoning - If the environment has multiple Fabrics, the additional HBA ports in the Host should have
separate unique Zones created in their respective Fabrics.
Q
QEMU - Also referred to as qemu-dm, which is the process name. The virtualization process which
allows full virtualization of a PC system within another PC system.
S
Server Pool - Logically an autonomous region that contains one or more physical Oracle VM Servers.
Presents a unified view of the storage where the virtual machines reside, and groups the users of these
virtual machines into a single community called a group, in which each user is a server pool member.
Single Initiator Multiple Target Zoning - Each fiber channel zone created should have a single initiator
(HBA port) and multiple targets (Storage Center front-end ports). This means that each HBA port needs
its own fiber channel zone containing itself and the Storage Center front-end ports. Zoning OVM hosts
by either port (commonly referred to as hard zoning) or WWN (commonly referred to as soft zoning) is
acceptable.
Sparse Copy - A clone of the type "sparse copy" is a disk image file of a physical disk, taking up only
the amount of space actually in use; not the full specified disk size.
See also Non-Sparse Copy.
T
Page 66
Thin Clone - A thin clone is a clone of a physical disk that takes up only the amount of disk space
actually in use; not the full specified disk size.
U
Utility Server - A component of Oracle VM Agent. An application that handles IO intensive operations
for virtual machines, server pools and servers, for example, copying, moving and renaming files. There
can be more than one Utility Server in a server pool. A physical server can perform as the master
Oracle VM Server, Utility Server and Virtual Machine Server simultaneously.
V
VIF - A virtual network interface for bridging network interfaces between domUs and dom0. When a
domU is started it is assigned a number. This number is used to bridge the network interface from ethn
to vifn.0.
Virtual disk - A file or set of files, usually on the host file system although it may also be a remote file
system, that appears as a physical disk drive to the guest operating system.
Virtual Machine - A guest operating system and the associated application software that runs within
Oracle VM Server. May be paravirtualized or hardware virtualized machines. Multiple virtual machines
can run on the same Oracle VM Server.
Virtual Machine Manager (VMM) See Hypervisor.
Virtual Machine Server - A component of Oracle VM Agent. An application which runs Oracle VM Server
virtual machines. It can start and stop virtual machines, and collect performance data for the host and
guest operating systems. Enables communication between the master Oracle VM Server, Utility Server
and Virtual Machine Servers. There can be more than one Virtual Machine Server in a server pool. A
physical server can perform as the master Oracle VM Server, Utility Server and Virtual Machine Server
simultaneously.
Virtual Machine Template - A template of a virtual machine. Contains basic configuration information
such as the number of CPUs, memory size, hard disk size, and network interface card (NIC). Create
virtual machines based on a virtual machine template using Oracle VM Manager.
VMM - See Hypervisor.
W
WWN Zoning - The zone only needs to contain the host HBA port and the Storage Center front-end
primary ports. In most cases, it is not necessary to include the Storage Center front-end reserve
ports because they are not used for volume mappings.
X
Xen - The Xen hypervisor is a small, lightweight, software virtual machine monitor, for x86-compatible
computers. The Xen hypervisor securely executes multiple virtual machines on one physical system.
Each virtual machine has its own guest operating system with almost native performance. The Xen
hypervisor was originally created by researchers at Cambridge University, and derived from work done
on the Linux kernel.
Page 67
Index B
Document references
Documents used and referenced in this document include:

Dell | Compellents Linux Best Practices
Dell | Compellents Live Volume Best Practices
Dell | Compellents Best Practices with Vsphere 5.x
http://docs.oracle.com/cd/E26996_01/e26529.pdf
Page 68

Dell Compellent OracleVM OVM Best Practices

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Dell Compellent OracleVM OVM Best Practices

Caricato da

Copyright:

Formati disponibili

Dell Compellent Storage Center

with Oracle OVM Virtual Machines

Menu items, dialog box titles, field names, keys

Mouse click required

User typing required

A note on Fibre Channel zoning

Host Bus Adapter Settings

Host Bus Adapter Queue Depth

Putting it all together:

Connecting your VM servers to the Dell Compellent array:

Legacy Primary/Reserved configuration

Virtual Ports configuration

Dell Compellent ports summary

Overview of Dell Compellent Storage Center features

Data Instant Replay (DIR)

Dynamic Capacity (Thin Provisioning)

Key points to consider:

Remote Instant Replay (Remote Replication)

Mapping Volumes to the Oracle VM Servers

Please see Linux best practices for more multipath detail.

Multipathing on Dell Compellent

Boot from SAN

Boot from SAN Configuration

Software-based server virtualization solutions for X86

Oracle VM Server is Xen Hypervisor.

Establishing Oracle VM Infrastructure

Implementation Tasks Overview

VM Server installation and VM Manager installation.

Fibre Channel Topology

Oracle VM 3.0.3 installation considerations

Oracle OVM Installation Tips

Physical and Logical Networks:

An example of Network and Fibre connections in an OVM Cluster.

Installing the Oracle VM software

Oracle VM Server Install

OVS 3.0 Eula

Remove all partitions and create a default Partition layout.

Oracle recommends to use a default partition layout.

Review Partition Layout

Boot Loader Configuration

Arrange Boot Drives

Oracle VM Management Interface

Select OK to connect the interface to a regular network.

Oracle VM Management LAN

IPv4 Configuration for eth0

Miscellaneous Network Settings

Time Zone Selection

Oracle VM Agent Password

Installing Oracle VM Manager

Steps to perform after Oracle VM Manager install

The Oracle VM Manger web GUI Interface.

Discover the two OVM Servers with the Discovery Tool.

Navigating in the home view Server Pools tab to a virtual machine.

Testing a Virtual Machine Migration

Manual migration of virtual machines, bringing up servers on another host.

Oracle VM Networking example of a large cluster utilizing many Server Pools.

Live Volume overview

Migrating the cluster with Dell Compellent Live Volume

Live Volume failover

# echo "- - -" > /sys/class/scsi_host/host0/scan

Live Volume replication attributes in Dell Compellents Enterprise Manager.

Documents used and referenced in this document include:

Potrebbero piacerti anche