Capacity Planner
Capacity Planner
Capacity Planner
You can find the most up-to-date technical documentation on our Web site at
http://www.vmware.com/support/
The VMware Web site also provides the latest product updates.
If you have comments about this documentation, submit your feedback to:
© 2007 VMware, Inc. All rights reserved. Protected by one or more of U.S. Patent Nos. 6,397,242, 6,496,847, 6,704,925,
6,711,672, 6,725,289, 6,735,601, 6,785,886, 6,789,156, 6,795,966, 6,880,022, 6,944,699, 6,961,806, 6,961,941, 7,069,413,
7,082,598, 7,089,377, 7,111,086, 7,111,145, 7,117,481, 7,149,843, 7,155,558, 7,222,221, 7,260,815, 7,260,820, 7,269,683,
7,275,136, 7,277,998, 7,277,999, 7,278,030, 7,281,102, and 7,290,253; patents pending.
VMware, the VMware “boxes” logo and design, Virtual SMP and VMotion are registered trademarks or trademarks of VMware,
Inc. in the United States and/or other jurisdictions. All other marks and names mentioned herein may be trademarks of their
respective companies.
VMware, Inc.
3401 Hillview Ave.
Palo Alto, CA 94304
www.vmware.com
2 VMware, Inc.
Contents
Preface 11
Intended Audience 11
Document Feedback 11
VMware Capacity Planner Documentation 11
Technical Support and Education Resources 12
1 Overview 13
VMware Capacity Planner Description 13
How Does VMware Capacity Planner Work? 14
Capacity Planner Collector 14
Capacity Planner Data Manager 15
Capacity Planner Information Warehouse 15
Capacity Planner Data Analyzer 15
Capacity Planner Dashboard 15
Tasks Performed by the Collector 16
Discovery Process 16
Inventory Data Collection 17
Performance Data Collection 17
Data Synchronization 17
About the Information Warehouse 18
Detection of Performance Anomalies 18
Data Upload and Processing 18
Network and Authentication Requirements 18
Network Connections 19
Capacity Planner API Overview 19
Discovery Process APIs 20
Domains 20
Servers 20
Inventory Collection API 20
Performance Collection 20
Moving Ahead with Capacity Planner 20
2 Planning an Assessment 21
Assessment Strategy 21
Basic Assessment Process 21
Project Road Map 22
Week 1 22
Pre-Assessment 22
Preinstallation 22
Installation 23
Data Collection Verification 23
Week 2 23
Collector Management 23
Data Analysis 23
Week 3 24
Collector Management 24
Data Analysis and Consolidation Scenarios 24
VMware, Inc. 3
Data Collection Guide
Week 4 24
Consolidation Reports 24
Assessment Report and Proposal 24
Uninstalling 24
3 Preinstallation Requirements 25
Personnel Requirements 25
VMware Capacity Planner Certification 25
Windows and Linux and Unix Troubleshooting 25
System Connections 26
Data Collection Methods 26
Network Troubleshooting 26
Data Collector System Requirements 26
Data Collector Host System 27
Host System Characteristics 27
Determining the Number of Collectors to Install 27
Accessing the Capacity Planner Download 27
Windows User Account for Collector Service 27
Target System Platforms 28
Considerations for UNIX and Linux Collection 29
Security, Network, and Firewall Requirements 29
Security Requirements 29
Trusted Domains 29
Account Privileges 29
Stored Account Information 30
Network Authentication 30
Firewall Considerations 30
Bandwidth 30
Preinstallation Checklist 31
Assessment Team 31
Collector Systems Information 31
Target Systems Information 31
Network Issues 32
4 VMware, Inc.
Contents
VMware, Inc. 5
Data Collection Guide
IP Scanning 68
Manually Adding Domains and Systems 68
Linux and UNIX Considerations 69
General Discovery Process 69
Discovering Domains 70
Discovering All Domains 70
Importing Domains and Systems 71
Manually Adding Domains 72
Deactivating Domains 72
Discovering Systems 73
Adding Systems 75
Adding a New System Manually to a Group 75
Adding Existing Systems to a Group 76
Working with Systems 76
Testing System Connections 76
Reviewing Collector Status Reports 77
Testing Windows System Connections 78
Working with Unknown Objects 78
Deactivating Systems 79
Deactivating a System in the Navigation Pane 79
Deactivating Objects from the Details Pane 79
Excluding Systems from Modules 80
Configuring the Collector for UNIX and Linux Collection 80
PuTTY Application 80
Validating Linux and UNIX Systems 80
Linux and UNIX Collection 81
Scripts Used for Linux and UNIX Collection 81
Manually Creating a crontab Entry When SSH Is Unavailable 82
Accessing the Discover Options 82
Setting the Discover Options 83
Setting the Group or Domain Discovery Options 83
Setting the System Discovery Options 83
Setting the IP Subnets Options 84
Setting the Active Directory Options 84
Setting the DNS Domains Options 84
Setting the Node Discovery Options 84
Scheduling Discovery Jobs 84
About Discovery Tuning 84
8 Inventory 87
Inventory Overview 87
Windows Management Instrumentation 87
Remote Registry and Perfmon 88
Secure Shell 88
Running Inventory Collection 88
Accessing the Inventory Options 88
Setting the Inventory Options 89
Setting the Inventory Collect Options 89
Setting the Inventory Control Options 89
Setting the Inventory Objects Options 90
Setting the Inventory Import Options 90
Setting the Inventory Export Options 90
Scheduling Inventory Collection 91
Inventory Reports 91
6 VMware, Inc.
Contents
VMware, Inc. 7
Data Collection Guide
8 VMware, Inc.
Contents
Glossary 175
Index 181
VMware, Inc. 9
Data Collection Guide
10 VMware, Inc.
Preface
The VMware Capacity Planner Data Collection Guide introduces Capacity Planner and provides information
about installing the client-side collector. After the collector is installed, this book guides you through steps
required for Capacity Planner to assess the client environment.
Information about using the Web interface to Capacity Planner (the Dashboard) is described in the VMware
Capacity Planner Dashboard Users Guide. The VMware Capacity Planner Dashboard Users Guide describes how to
view the company inventory and performance data. The Dashboard guide also describes how to analyze the
results to recommend consolidation and virtualization opportunities for the client.
Intended Audience
The intended audience for this book is primarily technical VMware personnel (consultants, Professional
Services Organization, training, support, and so on) and VMware partners who are responsible for installing
and using Capacity Planner in assessment engagements.
Document Feedback
If you have comments about this documentation, submit your feedback to:
Table 1. Abbreviations
Abbreviation Description
VMware, Inc. 11
Data Collection Guide
SAN Storage area network type datastore shared between managed hosts
tmplt Template
VC VirtualCenter
http://optimize.vmware.com
Customers with appropriate support contracts should use telephone support for the fastest response on
priority 1 issues. Go to http://www.vmware.com/support/phone_support.html.
Support Offerings
Find out how VMware support offerings can help meet your business needs. Go to
http://www.vmware.com/support/services.
12 VMware, Inc.
1
Overview 1
The following topics are described in this chapter:
Consolidation Estimate (CE) - A step-by-step workflow that quickly guides you through setting up and
conducting a new assessment. Within a week or so, you can present the customer with a report that
recommends two potential consolidation scenarios. This is typically enough information for the customer
to decide whether to conduct a full assessment.
Full Assessment - A thorough study of the customer’s IT environment conducted over a month or more.
After completion of a full assessment, you can present a detailed analysis of potential savings to the
customer available through consolidation and virtualization. You can also use findings from a
consolidation estimate by converting your CE assessment to a full assessment.
Capacity Planner can help you make informed decisions because it:
VMware, Inc. 13
Data Collection Guide
At the core of Capacity Planner is a unique Information Warehouse. The Information Warehouse contains
industry reference data that is used for comparative analysis and benchmarking to help guide validated server
consolidation and capacity optimization plans.
local Dashboard
collector
database collector host Web-based
analysis and
reporting
company’s
network
Capacity Planner quickly discovers systems within your client company’s IT environment, collects detailed
hardware and software inventory data, and collects key performance metrics from a target list of servers.
No agents are installed on any of the target systems. The target systems can be Windows, Linux, or UNIX
systems as described in Chapter 3, “Preinstallation Requirements,” on page 25. For Windows target systems,
collection is performed using standard Microsoft interfaces. For UNIX and Linux target systems, simple
scripts collect inventory and performance information using Secure Shell (SSH) connections.
The Collector has an advanced scheduler that allows it to manage jobs that are submitted to run discovery,
inventory, performance, and data synchronization modules at user-defined time intervals. For more
information, see “Tasks Performed by the Collector” on page 16.
On the average, each Collector system can monitor approximately 500 systems in the network. Multiple
Collector systems can be installed in the network, as deemed necessary. Chapter 3, “Preinstallation
Requirements,” on page 25 addresses the various factors that are used to determine how many Collector
systems will be required for an assessment.
14 VMware, Inc.
Chapter 1 Overview
Manages the process by which collected data is sent to the Information Warehouse
Provides detailed and summary views and reports on all discovered objects, collected inventory
information, and monitored performance data
Information collected from each company’s server environment is stored in the Information Warehouse, along
with aggregated information and metrics that serve as valuable industry benchmarks. You can use these
benchmarks for comparative analytics and scenario modeling.
For a more in-depth discussion of the Information Warehouse, see “About the Information Warehouse” on
page 18.
Aggregation
Trending
Benchmarking
Scenario modeling and what-if analysis help model and test various planning scenarios, including:
Virtualization
Hardware procurement
The Data Analyzer resides on the secured central host site along with the Information Warehouse.
Access your client company’s inventory and performance data in the Information Warehouse
VMware, Inc. 15
Data Collection Guide
Scenario modeling (including “what-if” analysis and analysis of hardware that your client might be
planning to buy but has not yet purchased)
To access the Dashboard, register with VMware to obtain a login ID and password. In addition, register the
client company to have collected data correlated correctly for your assessment. Registration is usually
completed with (or after) the purchase of assessment services from VMware. For more information, see
Chapter 3, “Preinstallation Requirements,” on page 25.
NOTE Use of the Capacity Planner Dashboard is documented in the VMware Capacity Planner Dashboard Users
Guide.
In addition to these main tasks, the Execute Command task is provided. This optional task lets you schedule
commands before or after one of the main tasks occurs. For example, before data synchronization, you can
perform a command to retrieve additional data polled from other collectors and put this additional data in
your outbox directory to upload with the new files. For more information, see “To specify a command” on
page 108.
Discovery Process
The discovery process uses a combination of:
Domains
Systems
Workgroups
Active Directory nodes
Even though a system or node in the company’s network is discovered, collection of inventory or performance
data from that system or node is not mandatory. Likewise, a node that is inventoried might not have
performance data collected from it. The number of discovered nodes will probably be greater than the number
of nodes that are inventoried or the number of nodes on which performance data is collected.
Domains can be excluded from any further processing. In this case, to proceed with inventory or performance
data collection from the nodes in an excluded domain, the domain must be re-enabled for that subsequent
16 VMware, Inc.
Chapter 1 Overview
processing. This setting is used because often a company does not want to inventory or collect performance
data on all of the domains in their network. For more information, see “Deactivating Domains” on page 72.
Domains and systems can also be added manually to the collector database. Importing domain and system
information from a comma-separated value (CSV) file is also an alternative method of getting the information
into the database.
The discover task is run once manually after the collector is installed. It might be run subsequently on a
schedule to determine whether additional nodes have appeared on the network and need to be discovered.
For more information about the discovery process, see Chapter 7, “Discovery with the Data Collector,” on
page 67.
Inventory data collection from Windows systems is usually accomplished through Windows Management
Instrumentation (WMI) first, and then by Registry connection, with the Collector connecting to each Windows
system through an administrator-level user account.
For UNIX and Linux systems, inventory data is collected by using Secure Shell (SSH). Permissions are
required. For more information, see “Considerations for UNIX and Linux Collection” on page 29.
Server information
Motherboard and chassis information
CPU (processor) information
RAM (memory) information
Information on drives and adapters
File system information
Network interface card information
Information on applications and services
Inventory is multithreaded. If the inventory task cannot retrieve information from a node with one method, it
tries alternative methods. The inventory process can be time onsuming. Messages on the progress of the
inventory for each system are written to a log file.
For more information about the inventory process, see Chapter 8, “Inventory,” on page 87.
As with the other main collector tasks, the performance task is usually performed manually the first time. It is
then scheduled to run repeatedly over a specified time period. The default scheduled interval is every hour.
The various aspects of performance data collection are discussed in Chapter 9, “Collecting Performance Data,”
on page 95.
Data Synchronization
Capacity Planner collects data with the inventory and performance collection tasks, makes the data
anonymous, and then uses the data synchronization task to transmit the data to the Data Analyzer for
processing and storage in the Information Warehouse. In addition to the standard security offered with data
anonymity, you can optionally mask server and domain names before data synchronization. In this case,
names are replaced with a pseudonym to further protect your customer’s privacy.
VMware, Inc. 17
Data Collection Guide
Data synchronization normally occurs automatically every hour, but the first execution is usually performed
manually after the first performance data collection is complete.
Scoring of collected performance data is accomplished by determining how many standard deviations each
statistic is from the industry average value stored in the Information Warehouse. The higher the number of
standard deviations, the further the statistic is from the average.
Anomaly detection allows industry performance averages to be used to provide a broad performance baseline.
Anomaly detection helps to discover problems such as:
By making performance tuning changes, you can help your client increase their company’s consolidation
opportunities and improve current service levels. Part of a successful optimization and consolidation strategy
is to improve a company’s processing environment so that they can experience these successes and feel
confident about proceeding in the assessment.
For example, a company’s Exchange server might support only 200 active users efficiently. However, the
Information Warehouse data might show that other Microsoft Exchange servers support 600 active users on a
server with similar processor, memory, and disk configurations. The lower capacity of your client company’s
Exchange server might indicate that a configuration or policy issue should be addressed. By limiting mailbox
sizes and distribution lists, using replication, a RAID configuration, or virus scanning technique, the client
might be able to increase the number of active users on that server.
For efficiency, the raw performance statistics and inventory files are uploaded in bulk to staging tables as
soon as they are collected.
Every 4 to 6 hours, the Data Analyzer processes the uploaded information in bulk. After this happens, you
can view most of the information through the Dashboard. A nightly job analyzes the loaded information.
Authorized and able to connect to all the servers that it is to manage using the protocols and ports outlined
in this document
18 VMware, Inc.
Chapter 1 Overview
Access to general purpose ports that Windows uses for most of its communications for file/print sharing
and authentication; they range from 135-139, and 445 (See Table 1-1 for more information about port
usage)
Account(s) with administrative rights to manage each server which can be a single account for all servers,
an account per domain, an account per server, or any combination
Network Connections
The collector system must be able to connect to all target systems by using specific protocols and ports. On
Windows systems, the general-purpose ports 135, 137, 138, 139, and 445 are used primarily for file and print
sharing and authentication. On UNIX and Linux systems, port 22 is required for Secure Shell (SSH).
The collector uses these ports first to pass credentials to a target system, and then to collect the actual data. Port
usage is described in Table 1-1.
22 TCP/UDP SSH Used for secure logins, file transfers, and port forwarding
135 TCP/UDP Loc-srv/epmap Microsoft DCE Locator service, also known as End-point DHCP Server
Mapper DNS Server
WINS Server
VMware, Inc. 19
Data Collection Guide
Domains
The discovery of domains is accomplished using the WNet API (mpr.dll) of Windows NT and Windows 2000.
This API uses the NetBIOS interface implemented by the Microsoft Lan Manager API to enumerate domains
and other beneficial properties.
Servers
The discovery of target Windows servers is accomplished using the Netapi32.dll. This API uses the NetBIOS
interface that the Microsoft Lan Manager API implements to enumerate servers of a certain class of system
such as Workstation, Server, SQL Server, Cluster, or IIS.
After the IPC$ resource is mapped to the remote target system, the appropriate inventory method for target
system platform is used to collect the information needed. The inventory methods include:
WMI—MS Windows 2000 / XP server data is usually collected using Windows Management
Instrumentation control (WMI). WMI calls are carried over the standard NetBIOS or Direct Connection
communications layers. These communications layers require ports 135 through 139 or 445 to be open.
Remote Registry and Perfmon—MS Windows NT 4.0 data is collected primarily with Remote Registry
calls and Perfmon (for memory) calls. These communications layers require ports 135 through 139 to be
open.
Performance Collection
Performance information is collected using one of two methods; one method is for Windows and the other is
Linux and UNIX. Windows performance collection allows you to add more counters. The performance
collection methods are:
Perfmon—Windows data is collected using the Perfmon API, requiring the establishment of an existing
authentication connection to any target system from which data is to be collected.
Perfmon calls are carried over the standard NetBIOS, requiring ports 135 through 139 or 445 to be open.
(The Perfmon APIs themselves use ports 137 through 139.) WMI can be turned off on systems.
SSH—Linux and UNIX data is collected using SSH only. An SSH session is established and scripts are
performed through that session or installed for later execution. If appropriate scripts are installed, the
data is returned using Secure Copy (SCP).
This requires the development of a thorough assessment plan, as described in Chapter 2, “Planning an
Assessment,” on page 21.
20 VMware, Inc.
2
Planning an Assessment 2
The following topics are described in this chapter:
“For tasks associated with or performed from the Dashboard, see the VMware Capacity Planner
Dashboard Users Guide.” on page 24
Assessment Strategy
A successful server consolidation assessment depends on understanding what you want to accomplish with
the assessment. There are numerous reasons for conducting a consolidation assessment. The company’s IT
infrastructure might be:
The company wants to evaluate what is happening on every system in their network (such as, which
systems are overused and which are underused) for confirmation of the perceived problems. With that
information, the company can then explore the effect of server consolidation on their IT environment.
Expanding the company network might be constrained by space, power capacity, or weight limitations of
the current data centers. It might be desirable to focus on assessing only older hardware and evaluate
whether the loads currently handled by those older servers can be supported more effectively on newer
state-of-the-art hardware, such as blade servers with virtualization.
If service levels for databases are a concern, you can limit the assessment to those database servers with
an emphasis on moving the load to larger, newer, more stable systems with greater memory resources.
The company wants to know how to use their current equipment more effectively, or if they can benefit
from removing two or more older systems that are not performing well and consolidating the functions
of those two systems onto a virtual machine. The assessment must focus on enough servers to show a
return that is large enough to cover the actual cost of the consolidation.
VMware, Inc. 21
Data Collection Guide
3 Integrating collected data with existing data in the Information Warehouse for analysis.
4 Analysis of company’s data with comparison to other industry data, scenario modelling, and
consolidation assessment.
After you understand the assessment strategy and company’s IT infrastructure, the first step using Capacity
Planner is installing the collector system—the collector and Data Manager—on one or more computers at the
company site. The number of installations of the collector system depends on the number of computers at the
company that are part of the assessment. In some cases, only a single installation is required.
After a collector system is in place, it can monitor and collect the necessary inventory and performance data
from a selected group of servers and other systems on the company’s network. Each collector system performs
this process for the group of systems that it is monitoring. The collection of inventory and performance data
is conducted over a specific period, as agreed upon with the company and Professional Services personnel
involved in the assessment engagement.
As the performance data is collected over time, it is transmitted on a regular basis (typically, once an hour) to
the Information Warehouse database, where it is processed and stored. As the data is collected.
A person in the company who is completely knowledgeable about, and has access to, the company’s IT
infrastructure.
A person who has in-depth knowledge of how to use Capacity Planner to perform the assessment, usually
a systems engineer consultant who specializes in IT consolidations.
Initially, the team needs to make an accurate determination of the company’s IT infrastructure and goals, to
understand the systems that are to be targeted for the assessment, and what is required to access those systems
to collect the necessary data.
An assessment should take three to four weeks at a minimum for a comprehensive view of the company’s
inventory and to collect significant data on how that inventory is performing. The following is a week-by-week
description of the typical assessment process.
Week 1
During the first week of the assessment, perform these tasks.
Pre-Assessment
Assess company objectives and requirements.
Preinstallation
Compile a profile of the company’s IT infrastructure (networks, firewalls, routers, servers, security issues,
operating system platforms, and so on).
Determine the number of collectors to install and set up (typically one per 500 systems to be assessed).
Through the Capacity Planner Dashboard, register the company in the Information Warehouse.
Determine locations of the systems that will host the collector and Data Manager.
22 VMware, Inc.
Chapter 2 Planning an Assessment
Chapter 3, “Preinstallation Requirements,” on page 25 presents a complete discussion of the information that
you must obtain, and the tasks that you must perform before the installation of the collector service and Data
Manager.
Installation
For each collector:
Download the latest version of Capacity Planner from the Dashboard Web site.
Test connections.
Use the Capacity Planner Dashboard to register the local collector database with the Information
Warehouse.
Week 2
During the second week of the assessment, perform the following tasks.
Collector Management
At the collector, perform weekly management tasks:
Data Analysis
From the Dashboard:
VMware, Inc. 23
Data Collection Guide
Group systems to identify organizational boundaries and assist in analysis of server loads and potential
consolidation.
Monitor progress.
Week 3
During the third week of the assessment, perform these tasks.
Collector Management
At the collector, perform weekly management tasks as described for Week 2.
Week 4
During the fourth week of the assessment, perform the following tasks.
Consolidation Reports
Near the end of the week (culminating a 30-day study), use the Dashboard to:
Uninstalling
Optional. Shut down and uninstall the collector installations.
For tasks associated with or performed from the Dashboard, see the VMware Capacity Planner Dashboard Users
Guide.
24 VMware, Inc.
3
Preinstallation Requirements 3
This chapter discusses what to do before installing the collector and Data Manager. The following topics are
described in this chapter:
Personnel Requirements
Ideally everyone on the assessment team is familiar with what Capacity Planner does, and what has to be done
to set up and start a collector for the assessment. In addition, the assessment team needs to be qualified in three
major areas:
VMware Capacity Planner Certification (not required if you are running CE assessments only)
Windows and Linux Troubleshooting
Network Troubleshooting
A VMware partner
Successful system connections (for more information see “System Connections” on page 26)
Access to data by means of one or more data collection methods (for more information, see “Data
Collection Methods” on page 26)
VMware, Inc. 25
Data Collection Guide
System Connections
To discover, inventory, and collect data from target systems, the collector must be able to access the target
systems with sufficient privileges that allow it to collect the data to be used for the assessment.
The assessment team must include at least one person who can identify and correct connection issues for the
collector. This person needs to identify which accounts are used to connect to the target systems, ensuring that
each account has the correct permissions for accessing data by one of the data collection methods.
For Linux and UNIX platforms, the data collection methods are:
NOTE Linux and UNIX data collection is done through shell scripts using standard Linux and UNIX utilities
to collect data from the target machines. Methods used to collect data are:
Local execution of shell scripts that are installed and executed on the target machine
Network Troubleshooting
The assessment team must also have one or more members who are familiar with the company’s network
architecture, from the dual perspective of security and performance. This expert must:
Be able to determine which domains exist and which systems belong to those domains.
Know the IT infrastructure (DNS, Active Directory, WINS, IP subnets, Active Directory organizational
units, and so on).
Know which ports can be used to gain access to any target systems behind firewalls, and be able to
identify whether it is necessary to install Capacity Planner on a system behind a firewall.
Know the general bandwidth of the network structure to optimize the number of collectors to be installed,
and where they are located within the network.
Be able to assist Windows, Linux, and UNIX experts in troubleshooting connection issues between the
collector and the target systems.
26 VMware, Inc.
Chapter 3 Preinstallation Requirements
If the collector is installed on a Windows XP system on which SP2 is installed, the Windows firewall
must be turned off.
If the collector is installed on a Windows 2003 machine, install the Win32_Product class, which is
required to enable the inventory feature to work correctly. Log on as an administrator, and open Add
or Remove Programs in the Control Panel. Under Management and Monitoring Tools, look for the
WMI Windows Installer Provider.
Connection to the Internet with network bandwidth of 20,000 bytes per second during collection
To access the Capacity Planner Dashboard from the collector, the following is also required:
You might have to install additional collectors to achieve optimal results of the assessment effort, but use as
few collectors as possible. The goal is to collect hourly performance data on every system within the project
scope. The realistic performance statistic depends on network throughput and server response, as well as the
presence of firewalls and routers in the collection environment.
VMware, Inc. 27
Data Collection Guide
the service to collect inventory or performance data.The user name, domain, and password for this account
are specified during the installation process. The user must have permissions to start a service on collector
host. You can do one of the following:
Set up the account and define the necessary permissions before starting the installation. The collector
service starts up automatically when the installation is complete.
Set up the account, but not the privileges. When you specify the account during the installation, the
installer checks for these privileges and notifies you. You can do one of the following:
Cancel the installation, grant the permissions to the user, and restart the installation from the
beginning. When the installation is finished, the service starts up.
Proceed with the installation and let it complete. The service does not start up at that time. Grant the
permissions to the user, and manually start the service before proceeding with setting up the
collector.
Defining the user and necessary permissions is described in Chapter 4, “Installing the Collector,” on page 33.
Windows 2003 Server (64-bit x86 / EM64T / AMD64) HP-UX 11.22 (PA-RISC)
SUSE Linux 10
SUSE Linux 9
SUSE Linux 8
RedHat Linux 9
RedHat Linux 8
AIX 5.1
AIX 5.2
AIX 5.3
28 VMware, Inc.
Chapter 3 Preinstallation Requirements
Security Requirements
For all of the target systems to be monitored, the collector must be able to connect to those systems using
specific protocols and ports. The collector uses these ports to pass to the target system the credentials that the
collector wants to use and to collect the actual data. The required ports are general-purpose communications
ports used for file and print sharing and authentication.
On Windows systems, the following ports are required to be open for a TCP connection: 135, 137 through 139,
and 445. On Linux and UNIX systems, port 22 is required for Secure Shell (SSH).
For information about how these ports are used, see “Network and Authentication Requirements” on page 18
in Chapter 1.
Trusted Domains
The collector host does not have to be in a trusted domain. However, the fully qualified account the collector
uses for connection to the target systems must be one of the following:
NOTE Trust relationships are transitive in Windows 2000 and later. This means that the trust relationship
extended to one domain is extended automatically to any other domain that domain trusts.
1 User A belongs to Domain A (and can connect to any system in Domain A).
5 Domain C trusts Domain A, and allows Domain A users to connect to Domain C systems.
Transitive trust relationships did not exist for Windows NT 4.0 (or any Windows system that preceded
Windows 2000). This means that Windows NT 4.0 accounts must be in the same domain as the collector or in
a directly-trusted domain.
Account Privileges
On Windows systems, the collector gathers inventory and performance data through WMI, the Registry, and
Perfmon. An account with local administrative rights to the target system is required to successfully collect the
information. The quickest way to collect information successfully is through a domain administrator account
VMware, Inc. 29
Data Collection Guide
or any account with administrator rights to all or most of the target systems. If no appropriate domain account
exists, a distinct account can be specified for each system after the collector discovers it.
For Linux or UNIX systems, data is collected using SSH scripts. The scripts are:
aog_setup.sh
aog_inv.sh
aog_inv_cron.sh
aog_perf.sh
aog_perf_cron.sh
aog_config_perf.sh
If you are not using the Data Manager to collect the data from the target Linux and UNIX machines, a root
administrator account is required to execute the shell scripts from the command line, and to copy the collected
data from the target machine to the import directory of the collector.
Network Authentication
Network authentication requires that file and print services and remote procedure calls (RPCs) are enabled on
target Windows systems. If the collector is using a domain account, you must be able to contact a domain
controller to authenticate the account.
Firewall Considerations
Firewalls protect the resources of private networks from unauthorized users from other networks. However,
firewalls can also prevent access by legitimate administrative users.
The collector lets you specify an account that allows connection to a target system or trusted domain. If you
cannot establish a connection directly between a target system (or a group of target systems) and the collector
or a trusted domain, the collector uses either a local account in the domain containing the target system or a
local account on the target system.
A firewall can affect a trust relationship. For example, a firewall might be open for the collector to connect to
a target system with a domain account, but the firewall might inhibit connection from that system to a trusted
domain to validate the provided account.
When a firewall is present within a network, it might be easier to install the collector and Data Manager on an
additional system that resides within the firewall.
Bandwidth
VMware recommends that you consider the effect of data collection on the network with regard to the rated
network bandwidth. In general, peaks of 20,000 bytes per second can be expected for each collector when
requesting inventory data or when collecting performance data. The performance data peak estimate
presumes collection of 500 performance counters per collection from four target systems. For more
information about performance counters, see Chapter 1, “Overview,” on page 13 and Appendix D, “Capacity
Planner Performance Counters,” on page 139.
30 VMware, Inc.
Chapter 3 Preinstallation Requirements
Preinstallation Checklist
Use this preinstallation checklist when you prepare for the initiation of an assessment engagement.
Assessment Team
Be familiar with Capacity Planner concepts and requirements for setting up a collector. (See “Personnel
Requirements” on page 25 and “Data Collector System Requirements” on page 26.)
Purchase appropriate services for the assessment and obtain instructions on downloading the installation
file. (See “Accessing the Capacity Planner Download” on page 27.)
System connections
Troubleshooting
System connections
Troubleshooting
Possess network expertise, especially for potential firewall and router issues.
You might want to optimize or adjust as discussed in “Determining the Number of Collectors to Install”
on page 27.
Determine the date of availability of host systems for installation of the collector.
Determine whether each host system meets minimum system requirements, as presented in “Data
Collector System Requirements” on page 26.
On each collector, obtain an administrator-level account to set up for the collector (for more information
see “Windows User Account for Collector Service” on page 27).
An administrator-level user account is also required to install the collector and Data Manager.
Number of target systems to be identified by discovery or from an import list ____ (See “Target System
Platforms” on page 28.)
VMware, Inc. 31
Data Collection Guide
DMZ
Be sure all target domains and target systems can be seen in the Network Neighborhood on the proposed
collector.
On each target Windows system, ensure that the following facilities are enabled:
Remote Registry
On each target UNIX or Linux system, ensure that port 22 is open and the sshd daemon is running and
available.
Network Issues
Be sure you have the following available:
User ID, password, and port numbers for the proxy server (if required)
Sufficient network bandwidth to collect performance data from all target systems within a one-hour
period
Bandwidth is determined by the number of systems targeted for full analysis and the number of
performance counters designated for the collection. For more information, see “Bandwidth” on page 30.
32 VMware, Inc.
4
Installation Overview
Figure 4-1 summarizes the collector installation process.
Uninstall Old
Version
(Optional - for
“clean” install)
Download
Capacity Planner
To create a clean installation on a system that has a previous version of the collector and existing data that you
no longer want to keep, follow the instructions in Appendix B, “Uninstalling the Data Collector,” on page 133
before running the setup wizard. If you have a version of the collector installed (regardless of the version) and
you want to retain the existing data, the setup wizard retains any existing data that you defined and collected.
Prior to beginning the installation process, you must know the user ID and password for the administrative
VMware, Inc. 33
Data Collection Guide
account that is specified as the collector service account in the setup wizard. For detailed information about
this account and the credentials required, see “About the Collector Service” on page 37.
Possible installation errors are described in “Possible Installation Problems” on page 40.
After you successfully install the collector, you can start it as described in “Starting the Data Manager” on
page 37.
1 Open a browser window and navigate to the Capacity Planner Web site at:
https://optimize.vmware.com
2 Log in to the Web site with the user name and ID that your VMware representative provides.
When you log in with a valid account, the Home page of the Capacity Planner Dashboard appears.
3 Click the Portal link at the top right of the Dashboard page.
4 On the Resources page, click the link for VMware Capacity Planner 2.6 Collector Build <number> where
<number> is the number of the current collector build. If multiple collector builds are listed, the highest
number is the latest version.
A File Download - Security Warning dialog box appears, prompting “Do you want to run or save this
file?”
5 Click Save, navigate to the place on your computer to download the installation file onto your system,
and click OK. VMware recommends that the installation file be downloaded and run off the local system
rather than over the network.
6 When you see the Download Complete message, click Run to begin the installation process.
If you have a previous installation of the collector on your system, read “Upgrading an Existing Installation”
on page 35
Is this a new assessment effort, with no consideration of previously collected data? For more information,
see “Performing a Clean Installation” on page 34.
Does the local database contain live data that you want to preserve for use in this assessment? If this
assessment is a continuation of one that was started with an earlier version of the collector, proceed with
“Upgrading an Existing Installation.”
When old installation is removed, proceed with “Installing a New Collector” on page 35. Even if the installer
program detects that an older version of Capacity Planner is installed and offers you the option of uninstalling
34 VMware, Inc.
Chapter 4 Installing the Collector
it, this method of uninstalling the application does not clean off the old data. Because the collector attempts to
update the old database, remove the old installation before you begin the new installation. For more
information, see Appendix B, “Uninstalling the Data Collector,” on page 133.
If the previous installation is the same version as the current installation, the installer uninstalls and
reinstalls the collector.
If the previous installation is an older version of Capacity Planner, the installer upgrades the collector.
Confirming the automatic upgrade retains everything that was set up for the collector in the previous
installation and allows you to continue your assessment with the newly installed software. See “Upgrading a
Version 2.x.x Installation” on page 36. If you retain the database from the previous installation, the Data
Manager does not automatically start the setup wizard for this new collector, and you need to manually run
the wizard after installing the new collector. For more information about the setup wizard, see Chapter 5,
“Setting Up the Collector,” on page 43.
1 If you are continuing the installation after just downloading, saving, and running the installation file, the
Security Warning dialog box appears.
Navigate to the downloaded installation file and double-click the file name.
4 Click Yes to accept the terms of the license agreement and click Next.
Click Next to accept the default destination folder (C:\Program Files\VMware\VMware Capacity
Planner) and proceed to Step 6 to specify the collector service account.
6 Enter the user name, password, and domain to be used as the service account for this collector.
The specific user, or a group it belongs to (such as, the Administrators group), has the privilege to log
on to Windows as a service.
If the collector service account does not meet these requirements, the service cannot start. For more
information about what can be entered in this dialog box and the results, see the “About the Collector
Service” on page 37.
NOTE The collector service account is not the same as the Global Connection Account that the collector
uses to query target machines for their system information. For information about the Global Connection
Account, see Chapter 5, “Setting Up the Collector,” on page 43.
VMware, Inc. 35
Data Collection Guide
The setup wizard authenticates the collector service account information. If the specified user cannot be
authenticated, a warning appears.
Click OK to return to the Service Account screen and enter correct user credentials.
When correct user credentials are entered, the wizard is ready to begin installation.
NOTE When the wizard begins the installation, it checks whether sufficient disk space is available for the
installation. If sufficient space is unavailable, the following message appears.
b Free up sufficient disk space, or determine another location for the destination folderthat has
sufficient available disk space. (For more information about system requirements for installation of
the collector, see Chapter 3, “Preinstallation Requirements,” on page 25.)
c Restart the installation procedure from the beginning (return to “Installing a New Collector” on
page 35).
As the installer finishes, it registers and starts the collector service. You can verify the state of the collector
service as described in Chapter 4, “About the Collector Service,” on page 37.
2 Navigate to the downloaded installation file and double-click the file name.
Because a previous installation of the same version of Capacity Planner is detected, you must confirm the
automatic upgrade.
When the wizard completes the configuration of the Windows Installer, the wizard indicates that it is
resuming the installation.
5 Click Next.
When the installation is completed, one of two InstallShield Wizard Complete screens appears.
If the upgrade required any modification to the Windows Registry, you are directed to restart your
computer.
If the Windows Registry was not modified, you do not have to restart your system.
36 VMware, Inc.
Chapter 4 Installing the Collector
6 In either Installation Wizard Completed screen, click Finish to exit the installer, indicating (if necessary)
whether you want to restart the system.
If you restart your computer, the collector service starts automatically. Verify the state of the collector service
as described in “About the Collector Service” on page 37.
From the Windows Start menu, choose (All) Programs>VMware>Capacity Planner Data Manager.
On your Windows desktop, double-click the VMware Capacity Planner Data Manager icon.
Proceed with Chapter 5, “Setting Up the Collector,” on page 43, to define the initial startup parameters for the
collector.
As with all Windows services, you can manage the collector service through the Windows Services facility as
described in “Viewing the Status of the Collector Service” on page 38. You can start and stop the service, if
necessary. (For more information, see Chapter 6, “Navigating the Data Manager,” on page 51 and Chapter 11,
“Management Tasks,” on page 105.)
With the account information you provide in the setup wizard, the installer attempts to authenticate the user.
If authentication fails, a message box appears.
You can then enter valid user credentials. If authentication succeeds, the installer then checks whether the
account has the privilege to log on to Windows as a service (on either the user or group level). If the account
does not have the privilege, the installer automatically assigns the privilege to the account (in the machine’s
local system policy) and proceeds with the installation.
The installer cannot register or start the service if any domain policy blocks it. In other words, a domain policy
that says “No service can be started on a domain account” does not allow the service to start on this domain.
If this occurs, you must specify another service account because the installer cannot alter the domain policy.
Table 4-1, “User Credentials and Installer Action,” on page 38 summarizes the variations of user credentials
that can be specified, and the resulting installer action.
VMware, Inc. 37
Data Collection Guide
Domain No Valid Valid Valid Passes, but service does not start.
NOTE If the machine on which the collector is to be installed is not in a domain (that is, it is in a workgroup),
you can leave the domain name field blank and the installer automatically adds the system name.
38 VMware, Inc.
Chapter 4 Installing the Collector
The Startup Type column shows that the service is Automatic, which means that it starts automatically
whenever Windows is started. The Status column shows that the service is Started. If the service is not
started, proceed with the next steps to verify service account designation.
4 To see more details about the service, right-click its name to see the Properties dialog box for the service.
5 To confirm that the service user is properly defined for the collector, click the Log On tab.
Verify that:
A specific account is defined as This account and the service is not running as a local system account.
The account that is defined is an administrator-level account with the privilege to log on as a service.
For more information, see Step 6 in “Installing a New Collector” on page 35.
VMware, Inc. 39
Data Collection Guide
NOTE The collector service does not start if the user name or password is invalid. This can happen if it
was specified incorrectly during installation, or if either have changed since the collector was installed.
You can redefine either or both of the user name and password for the service account in the dialog box
shown above.
For more information about managing Windows services, see Microsoft Windows Help .
You have sufficient available disk space for the Capacity Planner installation folder.
The Windows system on which the collector system is installed is properly updated.
The installation program might encounter Windows error 1931, although this is a rare occurrence. The error
can occur because the installer copies over some .dll files that are being protected by Windows. If this occurs,
click OK in the Windows error dialog box. The installation of Capacity Planner continues without any
problems. See additional information about error 1931 in your Windows online help.
In previous versions of Capacity Planner, the collector had to be upgraded manually when an update became
available. As of the 2.5 version, the updates can be made available to collectors installed in client sites
automatically. To use this feature, the collector machine must be able to connect to the Dashboard Web site.
Collector binaries
Third-party binaries
UNIX scripts
Product manifest file. This file contains information about all the components in the collector, including
name, location, and checksum information for each component.
When an automatic update occurs, the Data Manager shuts down and collector service restarts.
A verification process takes place after the download to verify that the checksums of the newly downloaded
components match those in the Dashboard manifest. If errors are found, they are logged into the collector log
file, the update is aborted, and the update failure is reported to the Dashboard.
40 VMware, Inc.
Chapter 4 Installing the Collector
1 Choose Admin>Options.
6 Select Download and install product updates to enable automatic updates to the collector. No other
check boxes should be selected.
For example, if data synchronization is scheduled for every hour of every day, every hour the Dashboard
compares the manifest checksum it has on file for this company with the manifest checksum in the collector.
If they match, nothing happens. The verification is performed an hour later, continuously, until a mismatch
occurs, at which time an update takes place. To limit the burden on the dashboard, add a data synchronization
job, only for automatic updates, that runs weekly.
VMware, Inc. 41
Data Collection Guide
3 Select Job Active, and enter a name and description for the weekly automatic collector data
synchronization.
7 Click OK.
8 Click OK again.
The data sync job, created specifically to perform updates of the collector, runs weekly on the day of the
week you specified.
42 VMware, Inc.
5
This chapter describes how to initially set up the collector including how to:
Use the setup wizard to define the main Global Connection Account. (Optional) Schedule the collection
tasks.
(Optional) Set global options for the connection account, file paths, and logging.
Basic navigation and descriptions of additional menu options in the Data Manager is provided in Chapter 6,
“Navigating the Data Manager,” on page 51.
To start the Capacity Planner Data Manager, select Programs>VMware>Capacity Planner Data Manager from
the Windows Start menu.
First-time installation: The setup wizard automatically starts to guide you through the initial setup of the
collector for the planned assessment. Proceed with “Using the Setup Wizard.”.
Already configured or upgrading a previous installation: The VMware Capacity Planner Manager
screen appears with all collector settings maintained from the previous installation. To examine or adjust
these settings, proceed with “Setting Global Options” on page 47.
VMware, Inc. 43
Data Collection Guide
Collect data—Choose whether you want to run a minimal assessment (Consolidation Estimate) or a full
assessment. If you have run a previous assessment, you can also elect to preserve existing configuration
settings.
Synchronization options—Configure settings that manage synchronization between the Data Manager
and the Dashboard.
Logging details—Specify number of log files to keep, maximum log file size, and level of logging detail.
Add accounts—Set up the Global Connection Account and other accounts as needed to access the target
machines within the scope of the assessment.
The following sections explain how the collector uses these settings. “Running the Setup Wizard,” describes
the procedure for setting these options.
The Global Connection Account allows the collector to authenticate against all or most of the systems and
domains detected by the initial discover task. The Global Connection Account must be an account that has
connection privileges on the domains and systems that are accessed for data collection.
However, because one account typically does not have administrative rights to every domain, server, or
machine in the network, you might need to identify other accounts for specific servers or domains to enable
connection to target systems for collection of inventory and performance data. “Setting Up Connection
Accounts” on page 105 provides additional information about adding connection accounts to accomplish the
inventory and performance tasks.
1 Click the Setup Wizard link on the Data Manager Home tab.
2 Choose the type of assessment you plan to run and click Next when you are finished. Choose:
Full—if you are planning to run a full consolidation assessment for at least 30 days
Preserve existing settings—only if you are running the setup wizard subsequent to a previous install
and want to keep custom settings that you entered previously
3 Choose synchronization options and click Next when you are finished.
Allow automatic software updates—enables automatic updates to the collector software when a
patch or hot fix becomes available
Enforce data privacy—disguises domain and host names before they are uploaded to the
Information Warehouse and Dashboard. For more information, see “Host and Domain Masking
(Data Privacy Feature)” on page 48.
4 Choose logging settings and click Next when you are finished. You can configure:
# of Backup Logs—determines the number of log files to retain; you must have at least one
44 VMware, Inc.
Chapter 5 Setting Up the Collector
5 Add user accounts needed to connect the collector host to the targeted systems. An account called the
Global Connection Account gives you access to most machines. Enter an account name, domain, and user
ID. Get this information from your contact at the customer site.
Click Add to set up additional accounts to access sequestered machines such as those behind a firewall.
You can also edit and delete accounts as needed. When creating an account to access UNIX networks,
ignore the Domain field.
Import systems from file—enables you to define the scope of the assessment by importing a list of
machines by host and domain name from a .csv file
Discover systems—configures the collector to find machines by discovering them in the customer’s
IT environment
7 Click Finish.
Your company must be registered in the Information Warehouse. One or more users must be granted access
to your company data with privileges at least at an Administrator level (within the Capacity Planner
Dashboard). See Chapter 3, “Preinstallation Requirements,” on page 25.
VMware, Inc. 45
Data Collection Guide
This process ensures that the correct data is stored in the Information Warehouse and presented for the
company in the Dashboard.
NOTE You must have Administrator privileges or greater in the Dashboard to register a collector.
To register a database ID
2 When the Register Collector dialog box appears, record the content of the Database ID field.
NOTE An easy way to perform this process is to select the entire contents of the Database ID field and
press Ctrl+C to copy it to the Windows clipboard.
3 Start an Internet Explorer browser window to connect to the Internet and navigate to the Capacity Planner
Web site:
https://optimize.vmware.com
5 Click the customer’s company name in the My Assessments table to display the Enterprise Dashboard.
All of the collector database IDs that are currently registered for the selected company are listed in this
table.
7 Click Add.
8 Enter the database ID for the collector database and click OK.
46 VMware, Inc.
Chapter 5 Setting Up the Collector
NOTE If you copied the database ID to the Windows clipboard when you retrieved it in the Data
Manager, press Ctrl+V to paste it directly into the Database ID field.
The Dashboard displays the Database ID Registration table again, showing the newly registered database ID.
For more information about the Database ID Registration table, as well as other objects presented in the
Capacity Planner Dashboard, see the VMware Capacity Planner Dashboard Users Guide.
Registered the local collector database with the Information Warehouse using the Capacity Planner
Dashboard.
You can proceed with your initial discovery of systems in your network. Guidance for setting up and
performing the Discover task is provided in Chapter 7, “Discovery with the Data Collector,” on page 67.
However, before you proceed to that chapter, read the information presented in Chapter 6, “Navigating the
Data Manager,” on page 51. This chapter familiarizes you with the layout of the Data Manager and enables
you to use it more effectively.
To adjust the collector options that you just defined, rerun the setup wizard, or follow the appropriate
procedures described in “Setting Global Options.”.
In the Capacity Planner Manager window, choose Admin>Options from the menu bar.
General Tab
The General tab, which is the default, defines the following options:
Logging
Other global options such as message window size and job history retention
See Table 5-1, “Where to Find Information About Logging and Other Global Settings,” on page 48 for
information about how you can set each of these options.
VMware, Inc. 47
Data Collection Guide
Table 5-1. Where to Find Information About Logging and Other Global Settings
Global Option Where Information is Located
Keep job history for x days Chapter 11, “Management Tasks,” on page 105
Ping Object before connection Chapter 12, “Troubleshooting Capacity Planner,” on page 115
If the customer has security concerns about this type of information leaving their company site, select Don’t
send host and domain names. In this case, an encrypted key for each host name is uploaded. The Dashboard
then generates a pseudonym for each name. Data for that host or domain is masked in the Dashboard under
its pseudonym, not the actual name. For example, a machine titled AcmeCoHoustonEmailSvr3 in the
company IT environment might be identified in the Dashboard under the pseudonym
B4C870B1-8259-6E45-BDB7-BCDC4DOF820D.
Capacity Planner provides a pseudonym key that allows you to generate a report in CSV format from the
Dashboard. The report maps the host and domain names to their pseudonyms. (In the Dashboard, choose the
Host and Domain Masking tab on Administration>Security Policy. Click Export Pseudonym Keys.)
After you revise all necessary options, exit the Options dialog box and proceed with discovery.
Root permissions are required for each UNIX or Linux system. Not having root permissions can result in
incomplete data collection while executing the scripts remotely. This is because only the root account can run
some of the utilities that the collector uses.
To enable remote connection to UNIX and Linux systems, Capacity Planner includes a third-party shareware
package named PuTTY. PuTTY is a Win32 Telnet and SSH client that allows you to connect remotely to a Linux
or UNIX system. The package is installed in the 3rdparty subfolder of the Capacity Planner installation folder
on the collector system. PuTTY allows you to use either private or automatically generated keys. Because SSH
does not allow automatic approval of generated keys, you must manually approve the connection to a Linux
or UNIX system one time before data collection can take place. See “Validating Linux and UNIX Systems” on
page 80.
PuTTY saves the necessary keys in the registry in the following location:
HKEY_CURRENT_USER\Software\SimonTatham\PuTTY\SshHostKeys
48 VMware, Inc.
Chapter 5 Setting Up the Collector
Because the keys are stored in HKEY_CURRENT_USER, they are unique to the logged-on user that effectively runs
the PuTTY application. The only way the collector can access those keys is if it is running as the same user as
PuTTY when it created the keys. This turns out to be the same user running the Data Manager.
To ensure that the collector can access the SSH keys, the collector must be set up with a collector service
account that matches the user who will be running the Manager.
NOTE Documentation on PuTTYis in its installation subfolder in the file putty.hlp. Double-click this file to
open the documentation in online help format. For specific questions about PuTTY, see the PuTTY
documentation, which is available from:
http://www.chiark.greenend.org.uk/~sgtatham/putty/docs.html
Capacity Planner collects performance and inventory data from active/passive and active/active clusters. In an
active/passive cluster, Capacity Planner ignores the logical IP address(es) assigned to the cluster and collects
data based on the server's physical IP address. Capacity Planner does not collect data from the passive node(s).
If servers in the assessment participate in an active/active cluster, Capacity Planner collects inventory and
performance data from those servers as if they were stand-alone machines.
VMware, Inc. 49
Data Collection Guide
50 VMware, Inc.
6
Configure and run the discover task to identify all of the systems in your network
Upload the collected data to the Information Warehouse (this process is called synchronization)
This chapter describes the key features of the Data Manager user interface. To use the Data Manager, you must
have:
Installed the Capacity Planner Data Manager and collector on a system in your network (or in a company’s
network) as instructed in Chapter 4, “Installing the Collector,” on page 33.
Set up the collector and registered the Database ID in the Information Warehouse as instructed in
Chapter 5, “Setting Up the Collector,” on page 43.
Basic Navigation
To start the Capacity Planner Data Manager, from the Windows Start menu, select
Programs>VMware>Capacity Planner Data Manager.
When the setup wizard finishes, the main Data Manager window appears.
VMware, Inc. 51
Data Collection Guide
Optionally, you might use the View Menu to also display the Message Pane. If the main window is closed with
the Message pane selected for view, this setting is remembered and the Message pane appears the next time
the Data Manager is started.
The status bar displays the job status of the collector, the status of the scheduler, and the logging level.
Displaying the status bar is optional and is determined by the setting on the View menu, as described in “View
Menu” on page 61.
NOTE Sometimes the logging level is not displayed if the Message Box is not displayed. (See “Message Pane”
on page 59 and “Setting Logging Global Options” on page 107.)
If the Data Manager window is minimized, the current job status is also displayed in marquee fashion in the
Windows task bar, allowing you to monitor job status while performing other non-Capacity Planner tasks on
the collector.
“Menu Structure” on page 61 provides an expanded discussion of these menus and their submenus.
52 VMware, Inc.
Chapter 6 Navigating the Data Manager
The icons allow you to perform the following functions (representing the icons from left to right):
Configure Options
The task-oriented icons correspond to the manual tasks that can be selected from the Tasks menu. The
Configure Options icon corresponds to the Options item on the Admin menu, and the last icon corresponds
to the item by the same name on the View Menu.
Navigation Pane
The Navigation pane appears on the left side of the main Data Manager window. Objects known to the
collector are displayed as an expandable tree structure in the Navigation pane. To see everything, open and
navigate to each branch.
NOTE The branches in the Navigation pane tree structure often have subordinate objects, even when they do
not indicate this. A branch might contain objects whenit does not have a plus (+) sign. The plus (+) and minus
(-) signs indicating that a branch contains objects do not appear until you open each branch and expand them
at least once.
Double-click any object in the navigation tree to expand the tree to display any subordinate objects.
The All Objects branch contains most of the objects with which you work, including all groups and systems.
The following example below shows the many objects that are included after discovery takes place.
VMware, Inc. 53
Data Collection Guide
Figure 6-8. Discovered Objects in the All Objects Branch of the Navigation Pane
Figure 6-9. Navigation Pane Showing Examples of Both Active and Inactive Objects
54 VMware, Inc.
Chapter 6 Navigating the Data Manager
The menu differs with each object. For example, if you might right-click one of the Group objects, you
might see the following menu.
If the object is active, the Active option in the right-click menu is selected. If the object is inactive, the
Active option in the menu is not selected.
The display in the Navigation pane changes to show only active items, and the toolbar light bulb icon appears
dim.
The display in the Navigation pane changes to show the inactive objects.
Enter the name of the object to find and click the search icon next to the box. If the object is found, the tree is
expanded to that point, and the object is highlighted and expanded (shown in Figure 6-10 for ada-win).
VMware, Inc. 55
Data Collection Guide
NOTE The contents and use of the pop-up menus are discussed in the remaining chapters of this book.
Collector Ownership
If the assessment is very large, you might be using multiple collectors. For example, if you are performing a
CE assessment and have 800 target systems in the scope of the assessment, you might collect data from 400
target systems using collector A, and 400 using collector B. When you are viewing The World, you can view
collector ownership to determine which collector is handling the objects you are viewing in the navigation
tree.
56 VMware, Inc.
Chapter 6 Navigating the Data Manager
3 If Owned is selected, the selected object belongs to this collector. For more information, see the Capacity
Planner Dashboard Users Guide topic describing load balancing.
Details Pane
The Details pane of the Data Manager screen appears to the right of the Navigation pane.
VMware, Inc. 57
Data Collection Guide
Home
Objects
Properties
Jobs
Job History
The operations that were already performed by the collector on that object (or on the entire “World”),
which might include information about current or completed jobs, basic discovery information, or
collected inventory data
For example, in Figure 6-12, the Job History tab of the Details pane displays the recent job history for the
collector.
Table 6-1 summarizes what you can expect to see on each tab.
Home Numbered links that guide you through a CE assessment. See the CE tutorial for more information.
Objects When a leaf with subordinate objects is selected on the navigation tree, this tab lists the subordinate
objects with detailed information about each object.
Inactive objects in the Details pane are dimmed as they are in the Navigation pane.
The Jobs and Job History tabs show the same information for all objects in the Navigation pane. The
information on these tabs changes only as jobs are run and finished.
58 VMware, Inc.
Chapter 6 Navigating the Data Manager
You can widen or narrow the entire Details pane by dragging its left border.
Message Pane
You can view the Message pane in the Data Manager window. Choose View>Message Box from the menu bar.
After the view is set to include the Message pane, the view remains that way for all subsequent starts of the
Data Manager until the view is changed to hide the Message pane. When the Message pane is selected for
viewing, it appears at the bottom of the Data Manager window.
Adjust the height of the Message pane by dragging the upper border of the pane. To see the full text of any
message line, point to the message line.
VMware, Inc. 59
Data Collection Guide
Right-click anywhere inside the Message pane to display a pop-up menu that allows you to copy the contents
of the pane to the Windows clipboard or to clear that portion of the window.
The messages displayed in the Message pane are the same messages that are logged in the collector’s log file.
This log file is located in the Capacity Planner installation folder. If the default destination folder was accepted
during the installation process, the path to the main log file is:
C:\Program Files\VMware\VMware CapacityPlanner\VMwareCP.log
The degree of detail in the logged messages is determined by the logging level that is defined as a global
option.
Figure 6-16. General Tab of the Options Dialog Box Showing Log Level Settings
The current setting of the Logging Level field appears in the status bar. See an example in Figure 6-3, “Status
Bar,” on page 52.
Other global options relevant to log management and the Message pane include:
For more information about setting these global options, see “Admin Menu” on page 64 and Chapter 11,
“Management Tasks,” on page 105.
To stop logging temporarily, select Pause Log at the top of the Message pane. Logging remains paused until
the box is deselected. VMware recommends that you pause logging only for a short time.
1 In the Navigation pane, navigate to the Objects tree and select either Objects or Systems.
2 Select the system (or multiple systems using Ctrl+click) to exclude from a collection task.
4 From the pop-up menu, choose the type of collection to exclude for this system.
60 VMware, Inc.
Chapter 6 Navigating the Data Manager
Menu Structure
This section describes the Data Manager menus.
File Menu
The File menu offers three options.
The Host/Domains tab allows you to specify a comma-separated value (.csv) file as a file
containing data to be imported. Click the View import systems template link for an explanation of
how the .csv file must be formatted.
For more information on importing host and domain information, see Chapter 7, “Discovery with the
Data Collector,” on page 67.
The Pseudonym Keys tab allows you to specify the name of a file created in the Dashboard that
contains data mapping keys for host/domain names that were disguised when exported to the
Information Warehouse. See “Setting Logging Global Options” on page 107 for information on
configuring data privacy options.
Export Host and Domain Keys starts an Export Keys dialog box that lets you name the .csv file in which
you will export the host and domain name mapping data. Click Save to save the data under the name and
location you selected.
View Menu
The View menu allows you to customize the Data Manager window.
You can choose whether to display the Toolbar, Status Bar, or Message Box, in addition to customizing what
is shown in the Navigation tree.
In the Navigation Tree Options menu, you can hide (or display) inactive objects, as well as specify categories
of objects to display.
VMware, Inc. 61
Data Collection Guide
Tasks Menu
The collector can perform the following generic tasks: discovery, inventory, performance, data
synchronization, and command execution. These tasks can be run at any time as manual tasks.
Some of these tasks need to be performed at specific times and at regular intervals. You can define scheduled
jobs. By default, numerous scheduled jobs are defined at the time the collector is installed.
Tasks allows you to run either generic or scheduled tasks on demand. Run Manual Task allows you to run
all manual tasks, or just a specific task.
The Run Scheduled Jobs menu varies depending on the currently-defined scheduled jobs. For example, right
after the installation of the collector, the menu appears as shown below.
Information about manual and scheduled tasks is presented in the subsequent chapters on the corresponding
operations of discovery, inventory, performance data collection, and data synchronization.
Information about scheduled tasks is also provided in Chapter 11, “Management Tasks,” on page 105.
Reports
The Data Manager provides summary or detailed information on the data that is collected by the collector.
Each report is a plain table displayed in a new window when the report is requested.
Chapter 11, “Management Tasks,” on page 105 presents guidelines and additional information on the content
and use of the Data Manager reports.
Five report categories are available from the Reports menu of the Data Manager. They are Inventory
Descriptions Reports, Inventory Reports, Performance Reports, Collector Reports, and Systems Summary
Report.
62 VMware, Inc.
Chapter 6 Navigating the Data Manager
Inventory Reports
You can view collected inventory data in either Summary or Detailed Inventory reports.
For a discussion of the Inventory process and the collection of inventory data, see Chapter 8, “Inventory,” on
page 87.
Performance Reports
Performance reports allow you to view performance data that is collected. You can view a summary report of
all performance statistics or focus on a specific category of performance data.
VMware, Inc. 63
Data Collection Guide
Collector Reports
Collector reports help you stay aware of the current state of the collection process.
For more information on the content and use of the collector reports, see Chapter 7, “Reviewing Collector
Status Reports,” on page 77 .
Admin Menu
The Admin menu provides additional aspects of management of the collector.
Options
The Options item of the Admin menu displays the Options dialog box.
Figure 6-26. Options Dialog Box Selected from the Admin Menu
64 VMware, Inc.
Chapter 6 Navigating the Data Manager
This is one of the most important dialog boxes in the Data Manager and is described in Chapter 11,
“Management Tasks,” on page 105.
Collector Management
The Collector item on the Admin menu allows you to control the execution of the collector service from the
Data Manager.
You can stop, restart, or uninstall the collector using these menu choices.
Stop—Stops the collector service. A Stop Service dialog box prompts you to confirm your action. If you
click Yes, the collector service stops until you manually start it. The menu path changes to
Admin>Collector>Start. Choose Start when you are ready to run the service again.
Uninstall—Uninstalls the collector. The Uninstall Collector Service dialog box prompts you to confirm
your action. If you click Yes to continue, the collector is removed from the host machine. VMware
recommends that you remove the collector service when you finish an assessment and will not be
returning to the customer site.
Database Management
You can perform management functions on the local collector database through the Database submenu.
Figure 6-28. Database Management Options Available from the Admin Menu
Compact Database—Reclaims poorly used memory in the AOGCapPlan.mdb file. You must stop the
collector service before compacting the database. The Compact successful message appears in the
Message pane when the task is finished. Remember to start collector service again when compaction is
complete. See “Unsuccessful Database Compaction” on page 125 for troubleshooting tips.
Check Database—Performs an integrity check of the AOGCapPlan.mdb file. The Database check
completed message appears in the Message pane when the task is finished.
Purge Database—Clears collected data from the AOGCapPlan.mdb file. A Purge Database dialog box
appears, prompting you to verify that you want to continue the purge process. All data in the
AOGCapPlan.mdb file is destroyed if you click Yes to continue. There is no Undo command, so be sure you
want to continue before clicking Yes. See “Purge Database” on page 126.
Display Database ID—Enables you to obtain the ID for this local database, so that it can be registered in
the Information Warehouse. Registration of the Database ID is documented in Chapter 5, “Setting Up the
Collector,” on page 43.
Setup Wizard
The Run Setup Wizard option (at the bottom of the Admin menu) usually runs automatically when the Data
Manager is started after a new installation of the collector. It is described in Chapter 5, “Setting Up the
Collector,” on page 43.
VMware, Inc. 65
Data Collection Guide
66 VMware, Inc.
7
Discovery Overview
After installing and setting up the collector, identify all target systems to include in the assessment. Identify
target systems using one of the following three methods:
The Discovery process uses standard naming services to find systems. An IP Scan option allows you to define
subnet ranges and ports for scanning.
The Discovery process finds groups and domains first, then computers within the groups and domains.
Chapter 12, “Troubleshooting Capacity Planner,” on page 115 presents many common problems with the
discovery process and offers suggestions for eliminating or avoiding those problems.
VMware, Inc. 67
Data Collection Guide
This chapter provides a general description of each name service and a discussion of issues that might arise
when you try to discover systems through that service. “General Discovery Process” on page 69 provides
information about how to set up your discovery tasks with regard to these naming services.
To access the browser naming service, a broadcast method is used. The broadcast method is restricted to the
Class C IP subnet in which the collector is a member. It also tries the LMHOST files. Workgroups in Microsoft
networks use browser name resolution.
WINS can be set up by DHCP or it can be manually set up in the network card configuration. Microsoft allows
up to 12 WINS servers to be specified in a search list. If problems are encountered when trying to access the
WINS servers, it is usually because of incorrect network card settings or the presence of firewalls in the
network.
Active Directory
Microsoft Active Directory usually requires the requesting system (that is, the collector) to be a domain
member before Active Directory Service Interface (ADSI) queries can be used. If you get an error that indicates
a table is missing during discovery, it is most likely that the Active Directory is unreachable or is not allowing
your query.
DNS
The domain name service (DNS) is the primary naming service for the Internet. DNS translates Internet
addresses to IP addresses. Within many customers’ networks, DNS is used as a primary naming service to
resolve system and domain names to IP addresses.
DNS is set up similarly to WINS and is affected by the same problems. In many companies, the DNS server
can restrict listing systems in the domain to reduce network use and DNS queries. If a company does not allow
listing systems in a domain, an alternative method is needed for the discovery task.
IP Scanning
If all the naming services are failing, or you want to find systems that are not part of any naming service, use
IP scanning to discover systems on the network. “Discovering Domains” on page 70 discusses how IP
scanning can be specified for the discovery tasks.
Manually add a named domain or system, one at a time, through the navigation tree. See “Adding
Systems” on page 75.
Import multiple named domains or systems from a comma-separated value data file. For instructions on
how to import domains and systems, see “Importing Domains and Systems” on page 71.
68 VMware, Inc.
Chapter 7 Discovery with the Data Collector
After you follow the process to discover the systems in the client environment, the discovery task runs once
daily by default. Discovery runs daily because environments change frequently—systems might be removed,
added, or might be offline temporarily. By running discovery daily or at least frequently, you ensure that all
systems from which you want to collect inventory and performance data are included.
NOTE This default setting applies only to MS LanMan systems and not to UNIX and Linux systems. Adjust
the discovery settings for UNIX and Linux environments.
The recommended technique for discovering domains and systems is described in Table 7-1.
1. Run a special job to discover only the domains in the “Discovering Domains” on page 70
network.
2. Confirm that all targeted domains are discovered. “Navigation Pane” on page 53
Check the Message pane or log file to see if any errors
occurred to affect the outcome of the Discovery task.
3. Deactivate any domains that are not to be included in “Deactivating Domains” on page 72
further data collection or assessment efforts.
4. Manually import or add any domains that were not “Manually Adding Domains” on page 72 and
discovered. “Importing Domains and Systems” on page 71
5. Run a Discovery task to discover only systems, setting “Discovering Systems” on page 73
the discovery options as appropriate.
6. Confirm that all desired systems were discovered. “Navigation Pane” on page 53
Check the Message pane or log file to see if any errors
occurred to affect the outcome of the Discovery task.
7. If necessary, manually import any systems that were not “Adding Systems” on page 75 and
discovered. “Importing Domains and Systems” on page 71
8. Deactivate any systems that are not to be included in “Deactivating Systems” on page 79
further data collection or assessment efforts.
9. Validate connections for all systems. “Testing System Connections” on page 76 and
“Validating Linux and UNIX Systems” on
page 80
After you have discovered all targeted domains and systems, proceed to Chapter 8, “Inventory,” on page 87
to set up and perform inventory data collection.
NOTE You might find it helpful to increase your logging level before beginning the discovery process. By
raising the logging level to Light Debugging (level 4), you receive more detailed information about each phase
of the Discovery task. This information is logged and can be viewed as it happens in the Message pane of the
Data Manager window.
VMware, Inc. 69
Data Collection Guide
The Data Manager provides a number of ways for you to discover domains and systems using the Discovery
task. You can customize the Discovery task using the discovery options for both manual and scheduled
discovery as described in Appendix 7, “Setting the Discover Options,” on page 83.
There are several ways you can run the Discovery task:
Discovering Domains
You discover domains by:
1 From the Data Manager menu bar, choose Tasks > Run Scheduled Jobs > Run Manual - Discover
Domains:
2 Click the Jobs tab of the Details pane to view the job status.
Let the Discover Domains job run to completion. You can also monitor the progress of the job in the status
bar or in the Message Box pane. For example, if the job is run with the logging level set to level 4 (Light
Debugging), information appears in the Message Box pane.
When the job is finished, information about the completed job is recorded in the Job History tab:
3 In the Navigation pane, expand The World to show the All Objects leaf.
Double-click the Groups leaf to see the domains. The domains are grouped by their type.
70 VMware, Inc.
Chapter 7 Discovery with the Data Collector
Select the Groups leaf in the tree to view the list of domains in the Objects tab of the Details pane.
They are listed by name in alphabetical order.
The default domain discovery task discovered domains of the following types:
Active Directory
LAN Manager
where groupname is a name that matches a group type in the list below and is always the same. Similarly, all
systems in an import file should be the same type. Valid group and system types are described in Table 7-2.
Within the listed groups are the domains: Domain Name Service (DNS), Microsoft Active Directory, and
Microsoft LAN Manager. In the case where the systems in the import file already exist and you are importing
a group, the systems are not recreated but they are assigned to the group. This is another way to assign systems
to groups.
2 Either type the fully qualified filename or browse the file system for the file.
3 Select the type of systems that are listed in the import file.
VMware, Inc. 71
Data Collection Guide
5 Click Import.
1 In the Navigation pane, select the Groups branch under All Objects.
6 Click Add.
The Data Manager adds a new group for this domain in two places:
The specific domain branch; for example, if the domain is a Microsoft LAN Manager domain, the
Microsoft LAN Manager branch also shows the new domain
To add systems to the new domain, see “Adding Existing Systems to a Group” on page 76.
Deactivating Domains
After you discover or identify the domains in a network, you might want to prohibit one or more of them from
being included in subsequent collection tasks. To do so, mark the domains as inactive. You can deactivate a
domain from either the Navigation pane or the Details pane of the Data Manager window, using the latter
technique to deactivate multiple domains in a single operation.
Deactivating a domain stops discovery on that domain and stops data about that domain from being uploaded
to the Information Warehouse. However, if the domain has systems assigned to it, deactivating the domain
does not stop inventory or performance collection on systems that belong to that group. This is because
systems can belong to multiple groups. To stop collection on a system, you must explicitly select that system
and deactivate it as described in “Deactivating Systems” on page 79.
If a domain that was discovered is no longer active, you can deactivate it and it won’t be uploaded to the
Information Warehouse. However, it remains dimmed in the navigation list. Remove it from view by using the
Show/Hide objects icon.
72 VMware, Inc.
Chapter 7 Discovery with the Data Collector
To deactivate a domain
1 In the Navigation pane, expand the navigation tree to show the domain to deactivate.
In the Details pane, you can select multiple domains to deactivate by pressing the Control key. However,
you can deactivate multiple domains only when they are grouped together on the same level under the
same leaf of the navigation tree.
When Active is not checked, the domain is inactive and is not processed in any collection tasks.
NOTE Inactive objects are dimmed in the Navigation pane. You can hide or show them by clicking the
Show/Hide objects icon.
Discovering Systems
The purpose of discovering systems is to define the scope of the capacity planning assessment.
You can run the Discovery task at any time. For example, you might want to run the Discovery task if you
added a domain or know that changes were made in your client’s environment.
Manually discover all systems. See “To discover all systems using the default settings provided with the
Data Manager” on page 73.
Discover systems within an IP address range. See “To discover systems within an IP address range” on
page 74.)
Set up a scheduled discovery job to automatically discover all systems or systems that fit specific criteria.
See “Accessing the Discover Options” on page 82.
In addition to using the discovery tasks, you can also add systems from which you want to collect by:
Importing system information from a file (See “Importing Domains and Systems” on page 71.)
Manually adding a system one at a time (See “Adding Systems” on page 75.)
To discover all systems using the default settings provided with the Data Manager
1 Choose Tasks > Run Manual Tasks > Run Discover Task.
A new job appears in the Jobs tab of the Details pane. Scroll to the right to see additional columns and
data.
VMware, Inc. 73
Data Collection Guide
You can monitor the progress of the job in the status bar, on the Jobs tab, and in the Message pane. What
you can see in the Message pane depends on the log level setting. The Message pane displays the progress
of the discovery task, and the status bar displays the percentage complete. When the discovery job is
complete, the following message appears in the Message pane:
You can also view the status of this job and all others in the Job History tab of the Details pane.
2 When you see Discover:Complete as part of the message text, click Job History to view the status of the
job.
3 Confirm that systems you want are included by expanding the Systems branch to visually inspect the
systems that are discovered.
To limit the discovery to an IP address range, modify the options for the task so that it includes the IP address
range in its search for systems.
1 From the Data Manager menu bar, choose Admin > Options.
2 Select the Modules tab.
5 Under the IP Subnet Ranges box, click Add to define an IP subnet to be scanned by the Discover module.
7 Click Add.
74 VMware, Inc.
Chapter 7 Discovery with the Data Collector
The newly added subnet appears in the IP Subnet Ranges area of the screen.
8 Click OK.
If the discovery task is run again, it now includes systems that are contained in the defined IP subnet range.
Adding Systems
The Systems branch under All Objects is where all systems that are known to be in your environment are
listed. The systems listed here are either discovered through the discovery process, imported from a file (see
“Importing Domains and Systems” on page 71), or manually added one at a time. You can manually add a new
system without assigning it to a group, or you can add a new system to a group.
1 In the Navigation pane, display the Systems branch under All Objects.
5 Select any groups that you want this system to belong to.
6 Click Add.
The system is added to both the All Objects > Systems branch and to each group you selected.
Add a new system manually, one at a time, by specifying the system information.
Display the groups in any one of the other group branches indicated by the group icon.
VMware, Inc. 75
Data Collection Guide
2 Select a group.
3 Right-click the group name to add a system to and select Add System.
7 Click Add.
The system is added to both the All Objects > Systems branch and to each group you selected.
2 In the Edit System dialog box under Group Membership, check all groups to which this system belongs.
3 Click OK.
The Objects tab displays a list of all systems that are discovered or defined along with detailed
information about each one.
3 Select one (or multiple systems using Shift-click) and drag it to any group name in the Navigation
pane.
The group must not be expanded when you drag a system to the group. Systems can belong to multiple
groups. See “Importing Domains and Systems” on page 71.
Deactivating systems
76 VMware, Inc.
Chapter 7 Discovery with the Data Collector
Attempt to collect inventory data first. After running the inventory task, you can see which systems were
successful, which helps you focus on the systems that had problems with connectivity.
To view the collector reports, choose Reports > Collector and select a report. While viewing the collector
reports, you can:
Select a system and right click to display a context-sensitive menu of other tasks you can perform on this
system.
Click Set Account to change the account being used for accessing a system or group of systems.
VMware, Inc. 77
Data Collection Guide
The Discovery process cannot determine the object type in the following instances:
The object type is anything other than a UNIX, Linux, Microsoft Windows Workstation, Microsoft
Windows Server and Microsoft Windows Cluster system.
The object type is one of the recognized systems, but the Discovery process was unable to connect due to
an incorrect user account.
78 VMware, Inc.
Chapter 7 Discovery with the Data Collector
In both cases, the object is created in the Unknown branch of the All Objects tree in the Navigation pane.
To change the Unknown objects to System objects, visually review the objects in the Unknown branch. If you
know that any are system objects, convert them to the appropriate system type by right-clicking the object in
the Navigation pane and select Create System. This changes the object type from Unknown to System.
Deactivating Systems
After you discover or identify the systems in a network, you might want to prohibit one or more systems from
being included in subsequent collection tasks. Mark the appropriate systems as inactive. You can deactivate a
system from either the Navigation pane or the Details pane of the Data Manager window by using the Details
pane to deactivate multiple systems in a single operation.
Data collected for deactivated systems are not sent to the Information Warehouse. The collector behaves as if
the deactivated systems don’t exist. If the Information Warehouse already has a record of the systems and then
they’re deactivated, on the next Monday the monitored flag is set to off and four weeks later the system data
are excluded from the four-week summary in the Dashboard.
1 In the Navigation pane, expand the navigation tree to show the system to deactivate.
NOTE You can select multiple objects by pressing the Control key. However, you can deactivate multiple
objects only when they are grouped together on the same level under the same leaf of the navigation tree.
When the Active option is not selected, the object is inactive and is not processed in any collection tasks.
Inactive objects are dimmed in the Navigation pane. You can hide or show them by clicking the
Show/Hide objects icon.
To inactivate one or more objects from the Details pane of the Data Manager window
1 In the Navigation pane, expand the navigation tree to show the object or objects to deactivate.
2 Select the leaf of the tree that contains the objects to deactivate.
3 In the Details pane, select the Objects tab to display all the objects contained in the selected leaf.
4 When the list of subordinate objects appears, select the objects to be deactivated, using standard Windows
selection techniques of Shift-click and Ctrl-click to select multiple objects in the list.
This clears the Active status for all selected objects, inactivating those objects.
VMware, Inc. 79
Data Collection Guide
Exclude All Tasks—Excludes the system from both inventory and performance collection
PuTTY Application
The PuTTY application is a free SSH client that you can download to the collector if it is not automatically
downloaded during installation of Capacity Planner. To download PuTTY, go to the following Web site:
http://www.chiark.greenend.org.uk/~sgtatham/putty/
PuTTY executable files are stored in the Capacity Planner 3rdParty folder.
The PuTTY SSH keys are stored in the host Collector Registry.
NOTE Documentation on PuTTY can be found in its installation subfolder in the putty.hlp file.
Double-click this file to open the documentation in online help format. For specific questions about PuTTY,
see the PuTTY documentation, which is available from:
http://www.chiark.greenend.org.uk/~sgtatham/putty/docs.html
In the Navigation pane, expand the navigation tree to display the UNIX or Linux system.
In the Navigation pane, select a high-level leaf containing the UNIX or Linux system, and select the
system on the Objects tab of the Details pane.
80 VMware, Inc.
Chapter 7 Discovery with the Data Collector
You can select more than one UNIX or Linux system to process in parallel by selecting the multiple
systems on the Objects tab, using standard Windows selection methods.
2 Right-click the selected system (or systems) and choose Utilities > Unix Terminal.
3 An SSH session is started to each of selected target systems. If prompted by a security alert to verify the
SSH key, click Yes. Log in using a valid login ID and password.
If multiple target systems were selected, an SSH session starts for the next target system until all selected
systems are processed.
A fast way to close multiple open PuTTY windows is to open the Task Manager on the collector (using
Ctrl-Alt-Delete). Click on the PuTTY applications, select all the PuTTY clients, and click End Task.
After the trust relationship is established for each UNIX and Linux target system, connections to the target
systems are successful for collection.
Table 7-3 shows a comparison of the system information collected, depending on the data source of the
information.
To configure how scripts collect data from Linux and UNIX machines
2 You can either select Run scripts remotely, which does not install anything on the target machines, or
Install and run scripts locally on each target machine. The advantage of running scripts remotely is that
the process ensures that the most current scripts are running. The advantage of installing and running
scripts locally is that there is less overhead on the collector host, since the work is distributed on each
target machine being polled.
VMware, Inc. 81
Data Collection Guide
Manual discovery — Run discover at any time using the Tasks > Run Manual Tasks > Run Discover
Task command. The discover options are what you set as the general default settings.
Scheduled discovery — Schedule one or more discover jobs by using customized discover options that
you specify for each job.
Use the same Discover Options dialog box for setting up manual discovery and for scheduled discovery jobs
set up using the job scheduler. Accessing the Discover Options dialog box is different for manual and
scheduled discovery.
82 VMware, Inc.
Chapter 7 Discovery with the Data Collector
For information about each tab on the Discover Options dialog box, see “Setting the Discover Options” on
page 83.
3 Select a Discover job under the Scheduled - Discover module and click Modify.
For information about each tab on the Discover Options dialog box, see “Setting the Discover Options” on
page 83. For information about defining and scheduling discover jobs, see “Scheduling Discovery Jobs” on
page 84.
Figure 7-2. Group/Domain Discovery Tab on the Discover Options Dialog Box
VMware, Inc. 83
Data Collection Guide
Figure 7-3. System Discovery Tab on the Discover Options Dialog Box
To use a custom query that will discover only certain objects, select Use custom Active Directory Search. Then
enter an ADSI query string in the blank field.
Scheduled - Discover
The process for scheduling all jobs (discovery, inventory collection, performance, and data synchronization) is
the same. For more information on scheduling jobs, see “Jobs Tab” on page 108.
84 VMware, Inc.
Chapter 7 Discovery with the Data Collector
To configure the discovery tuning attribute for a discovery task that is run manually
3 Double-click Discover.
In the Max number of simultaneous systems field, click the up and down arrows to indicate the preferred
degree of parallelism.
Experiment with this setting using trial and error to find the optimal number of simultaneous discovery
tasks for the collector machine you are tuning.
5 Double-click Discover.
7 In the Max number of simultaneous systems field, click the up and down arrows to indicate the preferred
degree of parallelism.
VMware, Inc. 85
Data Collection Guide
86 VMware, Inc.
8
Inventory 8
The following topics are described in this chapter:
Inventory Overview
The Inventory job collects hardware and software information about systems identified by the Discover task
and activated for data collection. You can inventory Windows, Linux, and UNIX systems. By default,
inventory collection runs every 15 days. In most cases however, because the data does not change often, only
a single inventory is necessary.
Typical inventory data that is collected might include information about a system’s CPU, memory, physical
disks, network interface cards (NICs), services or daemons, shares, and file systems. Data is generated using
utilities that are provided with the target machine’s operating system.
VMware, Inc. 87
Data Collection Guide
Secure Shell
You can collect UNIX and Linux data only by using SSH. An SSH session is established and scripts are
executed through that session or installed for later execution. If the scripts are installed, the data is returned
using SCP.
After the collector is installed, a Global Connection account is set up for the collector to connect to target
systems. See Chapter 5, “Setting Up the Collector,” on page 43. If the Global Connection account has
administrative privileges to connect to a target system, the account can inventory information from that
system. If the Global Connection account cannot collect information from a target system, you can identify a
specific user account and password for that target system, and the collector can use that account to connect to
the system. Chapter 11, “Management Tasks,” on page 105 describes how to set up a specific user account and
password for a target system.
A new job is created to collect inventory on all the objects in this group.
Manual inventory collection — run inventory collection at any time. Select Tasks > Run Manual Tasks
> Run Inventory Task. You set the inventory collection options as the general default settings.
Scheduled inventory collection —schedule one or more inventory jobs by using customized inventory
options that you specify for each job.
You use the same Inventory Options dialog box for setting up manual inventory as well as for scheduled
inventory jobs set up using the job scheduler. However, accessing the Inventory Options dialog box is
different.
88 VMware, Inc.
Chapter 8 Inventory
1 From the Data Manager menu bar, choose Admin > Options.
1 From the Data Manager menu bar, choose Admin > Options.
3 Select an inventory job under the Scheduled - Inventory module and click Modify.
For information about defining and scheduling inventory collection jobs, see “Scheduling Inventory
Collection” on page 91.
Collection Max number of simultaneous systems — Specify the number of systems the collector gathers
Tuning inventory data from simultaneously. This value is 10. When collection for a system completes,
another system is collected from so that this number of systems is always being collected from
simultaneously. If the collector seems sluggish during inventory collection, adjust this number
accordingly.
VMware, Inc. 89
Data Collection Guide
Collection Specify the criteria for which systems are inventoried.Only inventory systems that have not been
Control successfully inventoried is the default. This lets you limit the systems that are inventoried to save
processing time and the amount of data uploaded to the Information Warehouse.
Other Options Specify other inventory options to control the systems inventoried:
Track inventory changes — When the inventory data for a system changes, the collector adds a new
record. This is the default. If not selected, the previous inventory data for a system is deleted when
the system in inventoried.
Set to ‘Exclude’ status after <number> consecutive failures — If inventory collection fails for a
system, the collector tries again the next day. This value sets the number of consecutive times that an
attempt is made. When a system or object is excluded, it appears in the Details pane on the Objects
tab. See “Excluding Systems from Collection” on page 60.
Check the box next to the data type and click OK or Apply.
The excluded inventory data is not collected for any system or object.
For a complete list of the specific inventory data that might be collected for each object type and system type,
see Appendix C, “Inventory Data,” on page 135.
Check the box next to the data type and click OK or Apply.
The excluded inventory data is not collected for any system or object.
For a complete list of the specific inventory data that might be collected for each data type and system type,
see Appendix C, “Inventory Data,” on page 135.
The import and export settings are used during data synchronization. The data synchronization module calls
the import and export routines for all modules.
90 VMware, Inc.
Chapter 8 Inventory
Exclude these Check the boxes in the list for any objects for which you do not want to export inventory data.
object types
when exporting
The 500 machine limit is for performance, not inventory collection, because timing can be adjusted to account
for higher numbers of machines. This is a recommendation and not an absolute limit.
Inventory Reports
You can run summary inventory reports and detailed inventory reports to view the collected inventory data.
Different information displays on the Detailed version. See Table 8-4 and Table 8-5 below for a
comparison.
Systems Host name, host pseudonym if the data privacy option is turned on, IP addresses, OS, OS
version, OS service pack, serial number, chassis make, chassis model, maximum RAM,
maximum CPUs, CPUs, CPU speed, RAM size, NICs, total NIC speed, drives, total drive
space, number of applications, and number of services.
CPUs Number of CPUs, manufacturer, description, model, rated speed, FSB, first cache, second
cache, and third cache.
Memory Install count, manufacture, model, type, size, speed, form factor, data width, and total
width.
Drive Adapters Number of adapters, manufacturer, model, and type.
VMware, Inc. 91
Data Collection Guide
Chassis Amount of RAM (in MB), manufacturer, model, CPU type, maximum # of CPUs, RAM
type, # of RAM slots, maximum RAM, and # PCI slots.
Operating Systems Install count, manufacturer, name, version, major, minor, service pack, and build.
Applications Install count, manufacturer, name, version, major, minor, service pack, and build.
System System name, installation locale, current locale, system partition, system boot directory,
system comments, and system type.
CPUs System name, slot, CPU ID, current speed, manufacturer, description, model, rated speed,
FSB, first cache, second cache, third cache, FP, HT, MMX, SSE, and SSE2.
Memory System, manufacturer, model, type, size, speed, form factor, data width, and total width.
Drive Adapters System name, manufacturer, model, type, serial number, and bus number.
Drives System name, adapter ID, device ID, manufacturer, model, interface type,and drive
space (GB).
NICs System name, MAC address, IP address, IP type, DHCP server, name server, DNS
domain, gateway, subnet mask, manufacturer, model, NIC type, and NIC speed (MB).
Chassis System name, serial number, manufacturer, model, BIOS make, number of CPU sockets,
number of RAM slots, RAM max (MB), and number of PCI slots.
Operating Systems System name, company system registered with, registration owner, product ID, install
date, install location, manufacturer, name, version, major, and minor.
Applications System name, company system registered with, registration owner, product ID, install
date, install location, install source, manufacturer, name, version, major, and minor.
Services Seystem name, status, startup mode, run account, display name, unique name, and
description.
Shares System name, net name, type, remark, permissions, maximum users, and path.
File Systems System name, path, provider name, type, size, freespace, drive type, media type, and
compressed.
92 VMware, Inc.
Chapter 8 Inventory
CPU Manufacturer, description, model, rated speed, FSB, first cache, second cache, and third
cache
Memory Manufacturer, model, type, size, speed, form factor, data width, and total width
Chassis Manufacturer, model, BIOS info, CPU max number, CPU socket format, RAM number
slots, RAM max, and PCI number slots
Operating System Type, manufacturer, name, version, major, minor, patch, and build
Network problems
Service failures
Authentication issues
Because the problems encountered are common to other Capacity Planner tasks, discussion of them and
suggested troubleshooting steps are presented in Chapter 12, “Troubleshooting Capacity Planner,” on
page 115.
Refreshing Inventory
Inventory is typically run every 15 days. If you run a manual inventory on one or more systems immediately
after another inventory job is completed, you do not receive any results. Instead, the job is shown in the Job
History tab as a failed job, and the following message appears in the Messages pane.
No systems found using current settings. Make sure that there are systems, they are not all
inactive, and not Excluded at all levels, and haven't been collected recently. You can force a
performance refresh by right-clicking on the system and select Force Refresh --> Refresh
[inventory|performance] so that it has a check mark.
As directed in the message, you can reset the internal collection indicators by right-clicking the objects and
selecting one of the Force Refresh commands.
VMware, Inc. 93
Data Collection Guide
94 VMware, Inc.
9
Performance Overview
The performance job collects statistical information from all the discovered systems. As with inventory
collection, the collector must be able to connect to every target system with privileges to collect the
performance data from that system. By default, performance collection runs hourly.
For Windows systems, data is collected using Perfmon. Perfmon calls are carried over the standard
NetBIOS, requiring ports 135 through 139 or 445 to be open. The Windows performance collection allows
users to add more counters.
For UNIX or Linux systems, data is collected using SSH only. An SSH session is established and scripts
are executed through that session or installed for later execution. If the scripts are installed, the data is
returned using SCP.
Manual performance collection — Run performance collection at any time. Select Tasks > Run Manual
Tasks > Run Performance Task. You set the performance collection options as the general default
settings.
Scheduled performance collection —Schedule one or more performance jobs by using customized
performance options that you specify for each job.
VMware, Inc. 95
Data Collection Guide
Use the same Performance Options dialog box to set up manual performance as well as for scheduled
performance jobs set up using the job scheduler. Accessing the Performance Options dialog box is different
for manual and scheduled jobs. See “Setting the Performance Options” on page 96 for information about each
tab on the Performance Options dialog box.
1 From the Data Manager menu bar, select Admin > Options.
To access the Performance Options for running scheduled job performance tasks
1 From the Data Manager menu bar, select Admin > Options.
3 Select a performance job under the Scheduled - Performance module and click Modify.
Averaging Specify how much weight each data point has when averaged using the techniques below. The
higher the weight, the less effect each data point has on the average; the lower the weight, the more
effect each data point has on the average.
Geometric mean weight — The weight of each data point when data is averaged using the
geometric mean. This averaging technique removes peaks and valleys and is popular for
benchmarking.
Arithmetic mean weight — The weight of each data point when data is averaged using the
arithmetic mean.
Other Max number of simultaneous systems — Specify the maximum number of systems that
performance is collected from at the same time. The larger the sample size, the less value each data
point has.
96 VMware, Inc.
Chapter 9 Collecting Performance Data
Number of Performance The number of performance samples taken each time performance data is collected
Samples from a system.
Increasing the performance collection interval or the number of samples reduces the number of systems that
can be collected from each cycle.
Each time performance data is collected from a system, the collector collects a sample, waits the specified
interval period, takes another sample, waits again, and so on. The total collection time is the performance
collection interval multiplied by the number of performance samples. A typical setting is three samples at
4-second intervals, which gives a 12-second total collection time for each machine.
For data collected using Perfmon APIs, performance averaging algorithms use collected samples to calculate
an average value for the hour that more accurately represents the activity for the entire hour than is achieved
using the sample-based calculation.
Each performance counter shows whether it is active, the class that it belongs to, the specific instance that it
applies to: an asterisk (*) means any instance, and the actual metric that is collected. The performance counters
are listed alphabetically by class. These settings are global, not confined to a single task.
VMware, Inc. 97
Data Collection Guide
1 On the Performance Options Counters tab, select a performance counter from the list.
1 On the Performance Options Counters tab, select a performance counter from the list.
2 Click Import.
3 Enter the qualified file name or browse to the file system for the file and click Start.
The Data Manager imports the performance counters from the file. The file must be a CSV file that lists
the performance counters in the following syntax:
For example:
1,Processor,% Processor Time,_Total,0
0,Memory, Available Bytes,,0,
0,Test,Test,Test,0,
The following table describes the counter properties in more detail.
Table 9-4. Format for Imported Performance Counters
Property Description
PERF_Metric The metric (data being analyzed) for the counter. Examples
are “Available Bytes” for Memory, or “% Processor Time”
for processor. See Appendix D, “Capacity Planner
Performance Counters,” on page 139 for more information.
PERF_Instance The value for the metric. For example, 0, 1, _Total, and so
on for “% Processor Time.”
2 Select either Use local computer counters or Select counters from computer and choose the computer
from the list.
6 (Optional) Click Explain to open a window that displays the definition of the selected counter.
98 VMware, Inc.
Chapter 9 Collecting Performance Data
7 Click Add to add the counter to the list in the Performance Options Counters tab.
Removing a performance counter deletes both the counter and the collected performance averages.
1 On the Performance Options Counters tab, select one or more performance counters to remove.
2 Click Remove.
CAUTION The following action resets the performance counters to the original default counters. There is no
warning message.
2 Click Defaults.
The performance counters reset to the original list, and the summary table and any added counters are deleted.
The collector multitasks and can perform inventory and the performance tasks simultaneously. In addition,
different jobs can run across different domains simultaneously. The process for scheduling all jobs is the same.
For more information on scheduling jobs, see “Jobs Tab” on page 108.
The 500 machine limit is for performance, not inventory collection and is only a recommendation. Because
timing can be adjusted to account for larger numbers of machines, this is a recommendation and not an
absolute limit.
Performance Reports
You can run performance reports to view the collected performance statistics.
All Statistics System name, class, instance, metric, last value, and average value.
CPU Statistics System name, class, object, metric, last value, and average value.
Processor Utilization System name, current value, last value, running average, arithmetic mean, geometric mean,
minimum, and maximum.
Processor Queue Length System name, current value, last value, running average, arithmetic mean, geometric mean,
minimum, and maximum.
Memory Utilization System name, current value, last value, running average, arithmetic mean, geometric mean,
minimum, and maximum.
Network Utilization System name, instance, current value, last value, running average, arithmetic mean,
geometric mean, minimum, and maximum.
VMware, Inc. 99
Data Collection Guide
Physical Disk Utilization System name, current value, last value, running average, arithmetic mean, geometric mean,
minimum, and maximum.
Logical Disk Size System name, drive, current value, last value, running average, arithmetic mean, geometric
mean, minimum, and maximum.
3 With a report displayed, sort the data by clicking on any of the column headers.
Data Synchronization 10
The Data Synchronization process sends the data from the collector system to the Information Warehouse for
analysis and performs other tasks needed to synchronize the data.
Information Warehouse
The Information Warehouse is a repository of billions of performance statistics. It receives data from all
companies performing assessments with Capacity Planner. This allows Capacity Planner to compare and
contrast recent performance statistics for particular hardware and software configurations across companies.
The data in the Information Warehouse is kept anonymous. A company can view only its own specific data in
the Dashboard. However, it can view statistics from all of the other companies without the company-specific
information. The Dashboard allows companies to compare the performance of servers in their environment
with similar servers in the industry to identify performance anomalies and performance targets.
Inventory and performance statistics are sent to the Information Warehouse for processing. Appendix C,
“Inventory Data,” on page 135 and Appendix D, “Capacity Planner Performance Counters,” on page 139 list
the specific types of data that are collected and sent to the Information Warehouse.
5 Download product updates if you selected Allow automatic software updates of the Data
Manager/collector on the Collector/Dashboard Synchronization Options screen in the setup wizard.
6 Import inventory and performance files from the import directory if you enabled Import collection data
for systems on the General tab of the Data Synchronization module.
The Data Synchronization job sends the inventory files and raw statistics to the Information Warehouse where
they are uploaded in bulk into staging tables and processed. After processing, the information can be viewed
in the Dashboard.
The Information Warehouse processes nightly and weekly jobs to analyze the collected information. Data from
these jobs populates the graphs and tables in the Capacity Planner Dashboard.
Manual data synchronization — Run data synchronization at any time. Select Tasks > Run Manual
Tasks > Run Data Synchronization Task. You set the data synchronization options as the general default
settings.
Scheduled data synchronization —Schedule one or more data synchronization jobs by using customized
data synchronization options that you specify for each job.
You use the same data synchronization Options dialog box to set up manual data synchronization as well as
for scheduled data synchronization jobs set up by using the job scheduler. However, accessing the data
synchronization Options dialog box is different.
To access the data synchronization Options for running manual data synchronization tasks
1 From the Data Manager menu bar, choose Admin > Options.
1 From the menu bar of the Data Manager window, select Admin > Options.
3 Select a job that contains a data synchronization task under the Scheduled - Data Synchronization
module and click Modify.
Export collection data for systems Writes inventory data to the outbox directory.
Get collector settings changes from Dashboard Retrieves the collector settings from the Dashboard.
Download and install product updates Automatically downloads and installs product updates.
The data that is exported depends on the selections you choose on the Export tab of the Inventory Options
dialog box.
2 Click the Inventory module to display the Inventory Options dialog box.
1 Go to https://optimize.vmware.com.
2 When you purchase the assessment services from VMware, an account is set up in your partner company:
c Click OK.
Management Tasks 11
The following topics are described in this chapter:
Connection Accounts — Working with the connection accounts to ensure that you can connect to the
domains and systems to collect data from.
Admin Options — Working with the Admin Options dialog box, which is the main user interface tool
used to configure the collection modules, jobs, and scheduling, as well as connection to the Dashboard.
Reports — Understanding the content of the reports provided with the Data Manager.
If the Global Connection account you initially defined in the setup wizard is not sufficient to collect
information from a target system, you can identify a specific local connection account for that target system,
and the collector uses that account to connect to the system instead of the Global Connection account.
You can define a local connection account for a single target system, or you can define a local connection
account to be used for a group of two or more target systems.
1 With the World > All Objects > Systems leaf selected in the navigation tree, do one of the following:
b Select one or more systems in the Objects tab of the Details pane, using Ctrl+click to select
multiple systems.
The account information appears after the Use global connection account field.
3 To specify a different account, select the Use selected account to specify a new local connection account.
You can select an existing account from the displayed list or create a new account.
a To select an existing account, click the account name and click OK.
c Enter the necessary information for the user account that is to be used to connect to the target
systems.
All data entry fields in the Create New Account dialog box are required except the Domain
specification. Generate a list of known domains from which to select by clicking Browse next to
Domain.
d Click OK.
4 Click OK.
The connection account (system specific or global) that is used for each active system is listed on the Objects
tab of the Details pane when the appropriate leaf is selected in the navigation tree.
2 Select the Objects tab in the Details pane and scroll to the right to view the Account column.
General Tab
The first tab of the Options dialog box is the General tab. General options specify configurations such as
Global Connection account information, data privacy options, and how logging and other information is
presented. The options on this tab are discussed in the following sections.
3 In the Create New Account dialog box, enter the necessary information for the user account that is to be
used to connect to the target system.
All data entry fields in the Create New Account dialog box are required, except the domain specification.
To specify the domain, generate a list of known domains to select from by clicking the browse icon next
to Domain.
4 Click OK.
The new account becomes the global account in place of the account that was defined previously.
To select a Global Connection Account as the active account for the collector
On the Global Options tab, the Account field displays the active Global Connection account.
2 Click the down arrow of the Account field to display a drop-down menu of currently defined accounts.
If the active Global Connection Account cannot be used to connect to specific target systems, define connection
accounts for those systems, as described in “Setting Up Connection Accounts” on page 105.
1 On the General tab, select the account as the active Global Connection Account.
2 Click Edit .
2 In the General tab, select the account as the active Global Connection account.
3 Click Delete.
4 Click Yes.
5 Activate one of the remaining accounts as the Global Connection account for the collector as described in
“To select a Global Connection Account as the active account for the collector” on page 107.
The # of Backup Logs option on the General tab specifies the maximum number of backup log files that are
created. By default, up to three backup logs are saved in the main Capacity Planner base installation directory.
When you set this option, consider the size of each log file and the amount of available disk space on the
collector system. Saving backup log data can be valuable for later assessment of collector activity.
The Log Level option on the General tab determines the amount of detail that the collector logs for the tasks
it performs. The collector records exactly what happens while the collector is running at the level specified
from a minimum level of no information to a maximum level where every detail is logged. The log provides
valuable diagnostic information for tracking activity and analyzing problems.
As you move the slider to the right, the logging level increases. Six levels of logging are available. The current
log level is shown just above the slider to the right. As the logging level increases the amount of information
logged, the log file fills up more quickly.
The Message Window Size option on the General tab refers to the maximum number of messages that are
displayed (with scrolling) in the message box pane at the bottom of in the Data Manager main window. The
default message window size is 500 lines.
By default, the message box pane is not displayed. Choose View>Message Box to display the message box
pane. The messages displayed in the message window are saved in the collector’s log files and can be viewed
with any text editor.
The Max # Concurrent Jobs option on the General tab defines the maximum number of concurrent jobs that
the collector allows. The number of jobs that are needed to complete the task and the maximum number of
concurrent jobs that the collector allows affect the performance of the collector and the duration of any task.
The Keep job history for x days option on the General tab determines the length of timethat jobs are listed in
the Job History tab of the Details pane in the Data Manager. The default value of 30 days is usually sufficient
to track all jobs the collector performs.
At any time, you can right-click one or more jobs listed in the Job History tab and delete the archived job using
the pop-up menu. You can also completely clear the list of archived jobs or save the list to a comma-separated
value (.csv) file.
The Ping Object before connection check box on the General tab specifies whether the target system is pinged
before a connection attempt is made. If the collector first pings a target system before trying to connect to it, it
can determine more quickly if the system is available. The Ping Object before connection option is not
selected by default.
The purpose for the Ping Object before connection option is to reduce collection times. When the Data
Collector tries to collect inventory or performance data from a target system, it needs to connect to the system.
If the target system is not available, the connection attempt times out. If this occurs frequently, it can increase
the collection times.
CAUTION For security reasons, some customer networks disable acknowledgement (or receipt of) a ping
operation for some or all of their systems. This does not prohibit the collector from connecting to those systems.
However, the ping attempt fails, causing the collector to bypass the system. If you are not sure whether all
potential target systems can respond to the ping attempt, deselect this option.
Modules Tab
The Modules tab of the Options dialog box lists the installed modules of the collector. For each module, the
tab shows whether the module is active and which version and build number is installed.
When you select one of the modules and click Settings, you display the Options dialog box for the selected
module. Use these options to customize the default behavior of each module. The specified settings are used
when you run the manual tasks through by choosing Tasks > Run Manual Tasks. Likewise, when you
right-click an object in the Navigation pane and choose a Tasks option, you run a manual task using these
option settings.
The Execute Command module allows you to specify a command to run globally.
To specify a command
4 Click Apply.
Jobs Tab
The Jobs tab on the Options dialog box lets you view all defined jobs and their schedules. The predefined jobs
are listed with their schedules. You can modify these or add your own custom jobs. Each job contains at least
one task that runs a Capacity Planner module, but it can contain many tasks that each run a Capacity Planner
module.
The Modules tab allows you to set default options. These options apply to all manual tasks that are run.
Manual tasks run only when you manually execute them. For a valid assessment of your company’s network,
run many of these tasks on a regular cycle over a specified period of time. In addition, you can define a custom
task to perform only a certain subset of options for a particular job.
When the collector is installed, scheduled jobs are defined for immediate use. The Jobs tab of the Options
dialog box lists these tasks.
The following scheduled jobs are defined but are in an inactive state:
Manual – Discover Systems (to discover only servers and other non-workstation nodes on the network)
You must run these last jobs manually by using the Tasks > Run Scheduled Jobs or activate them on the
Schedules tab to have them run according to a schedule.
The names of these jobs reinforces the intended purpose of running these jobs manually before activating the
scheduler and starting the cycle of the other scheduled jobs.
When the Data Collector is installed, the job scheduler is initially suspended so that the predefined scheduled
jobs do not start to run before you are ready. Before activating the job scheduler, discover your domains by
running the Manual - Discover Domains job manually, and then the Manual - Discover Systems job. See
Chapter 7, “Discovery with the Data Collector,” on page 67.
Deselect Suspend Scheduler to start the Scheduler and run all active scheduled jobs as per their
defined schedules.
Select Suspend Scheduler to stop the Scheduler from running any active scheduled jobs.
Capacity Planner has predefined jobs that are scheduled to run every day at the frequencies described in
Table 11-1.
You can view and modify jobs schedule at any time. See “To define a scheduled job or customize a previously
defined job” on page 110.
You can run any scheduled job manually, regardless of the state of the job scheduler (active or suspended) or
the defined schedule for the job.
From the Data Manager menu bar, select Tasks > Runs Scheduled Jobs > Run <desired job>.
This overrides the state and scheduling of the selected job and runs it immediately. You can run all jobs on
demand by this method.
This functionality is useful to run custom or specialized jobs such as the default jobs to discover only domains
or systems.
You can define scheduled jobs or modify the predefined jobs for your specific environment and requirements.
The task types of the scheduled jobs are execute command, discovery, inventory, performance, and data
synchronization. The process for scheduling jobs is the same for all of these types.
When a new scheduled job is defined, it is included on the list of scheduled jobs that you can see in the Tasks
menu. You can also define a schedule for this job so that it is run by the job scheduler at the times and frequency
that you designate.
If customizing an existing job, select the job in the list and double-click it or click Modify.
Either the Modify Job or the Add Job dialog box is displayed. They are identical, except that the Modify
Job dialog box displays the information that defines the selected job. Each of the tabs in the Modify Job
dialog box is described in the following sections.
The General tab in the Modify Job dialog box is where you activate or deactivate the job as well as specify
the job name and description.
4 On the General tab, enter or modify the name and description of the job.
Do not activate the job until you finish defining it. If you are modifying an existing job, deactivate it by
deselecting Job Active. See “To activate or suspend the job scheduler” on page 109.
5 In the Modify Job dialog box, click the Schedule tab and define the timing and schedule of the job.
The Schedule tab shows you the scheduling method, duration, daily frequency, and recurrence for this
job.
6 Set the options for scheduling this job.
Scheduling Method Submit to Submit – The job begins at the specified time interval following the
submission of the previous job. For example, if you specify daily recurrence with
a frequency of one hour, the job begins exactly one hour after the submission of the
previous job.
Start to Submit – The job begins at the specified time interval following the start
of the previous job. For example, if you specify daily recurrence with a frequency
of one hour, the job begins exactly one hour after the start of the previous job.
End to Submit – The job begins as of the specified time interval following the end
of the previous job. For example, if you specify daily recurrence with a frequency
of one hour, the job begins exactly one hour after the previous job ends.
Duration Start Time – Date and time the job schedule begins.
Daily Frequency Run every – Frequency this job runs each day in number of minutes, hours, or
days.
7 In the Modify Job dialog box, click the Tasks tab to view and add to or modify the job task options.
The Tasks tab displays the tasks for this job to be perform, the criteria by which processing can continue,
and the time limits for processing.
9 Select a module, specify the timeout settings, specify the continuation settings (you can choose more than
one), then click OK.
The defaults are described in the following table.
Task Timeout If the task takes longer than the set time to complete, timeout
occurs and the task is terminated.
Object Timeout If collection for a particular system takes longer than the
specified time, object timeout occurs for that system and the
task is terminated for that system. Collection continues for
other systems.
Idle Timeout Timeout and then a restart occurs when any task is idle for
more than the specified time.
Continue on to next task if this status is Cancelled – The next task begins if the previous task is
cancelled.
Success – The next task begins if the previous task is
successful.
The request for performance data can time out on some target systems. To correct this, increase the
timeout for performance collection at the object level.
If the collector experiences idle time of approximately 20 minutes or more, you might see OpenDatabase
errors in the log (which means the database on the collector host machine cannot open and accept data).
An example of this type of message is: “08/16/2006 10:44:57:0:3880: OpenDatabase: All attempts to open a
database, failed. This problem might be due to high activity by the collector. Reduce the number of
systems collected at the same time if this is the case.” To correct the problem, reduce the number of
systems collected at the same time, as suggested in the log message.
10 (Optional) On the Tasks tab, click Objects to apply this job to a specific set of objects.
11 (Optional) On the Tasks tab, click Settings to modify the settings for the selected module.
12 Repeat steps 8 through 11 to add or modify additional tasks for this job.
View the job history statics for the job in the Statistics tab in the Modify Job dialog box.
15 Select the job from the Jobs tab of the Admin Options dialog box.
17 On the General tab of the Modify Job screen, select Job Active.
18 If you created a new scheduled job, verify that the job was created.
You can customize any of the tasks that run as part of a job.
6 In the Options dialog box, specify a set of objects to which a job task applies.
9 Select Job Active to activate this new job for the job scheduler. to handle
2 Check the type of module task to add, specify the timeout settings, and click OK.
3 Select the task and modify the task objects and settings.
If you do not want to run a job, you can delete it. You can also deactivate a job in case you ever want to run it
again.
To delete a job
CAUTION If you select a task listed under a job name, the entire job is deleted, not just the selected task.
To delete a task
5 Select the task that you want to delete, and click Delete.
Collection Tab
In the Options dialog box, the Collection Settings tab allows you to specify the manner in which UNIX or
Linux scripts are run on non-Windows systems.
The scripts run locally on the collector host and then access the remote target systems. You don’t have to
install anything on the target systems. You must have SSH setup enabled and the proper permissions to
connect to each target system.
Install your scripts in a specified location on each of the UNIX or Linux target systems.
If you choose to install the scripts, you can set them up for scheduled execution by using the crontab
command.
Database file for collected data—A local Microsoft Access database file named AOGCapPlan.mdb in
which collected data is stored.
Base installation directory—A directory in which all VMware Capacity Planner directories and files are
stored.
Import directory—The directory in which collected data is stored if data is collected from Linux and
UNIX target machines by installing collection scripts on each target machine as described in “Collection
Tab” on page 113 and “Running Scripts Manually” on page 171.
Outbox directory—A directory in which data that is ready to be uploaded to the Information Warehouse
or exported to a .csv file is stored.
Archive directory—You can archive the files on the collector host if you choose Archive files after data
processing. The location of the Archive folder appears in the field below Archive.
You can view and modify the default name and location of these file paths.
The File Paths section in the File Paths tab displays the current file paths for the database and base
installation directory, the import and outbox directories, and the archive directory.
If you make this change, validate the database that you are redirecting the collector to.
2 To change the location of these files or directories, click Browse and make a note of the new location.
CAUTION VMware does not recommend that you change these path names unless directed to do so by a
VMware Capacity Planner Certified Provider.
Connection Tab
The Connection tab of the Options dialog box enables the connection to the Information Warehouse by
identifying and specifying the URL of the Information Warehouse and the proxy settings, if any, that are
needed to establish that connection through your network structure.
Reports
The Data Manager provides detailed reportsthat show the data the collector gathers. Each report is a table
displayed in a new window when the report is requested.
Often many columns are out of view when you first open a report. Scroll to the right to see the hidden columns
or adjust the report display window size. Initially, the data is listed in the chronological order in which it was
collected. You can sort the data in ascending or descending order by any column by clicking the column
header.
Every report that the Data Manager provides can be saved to a file on the collector system by clicking Export
at the bottom of the report window. A report is saved as a CSV file and can be imported into Microsoft Excel
or other reporting or analysis tools.
The reports are not intended to provide information for assessment of the company’s network and IT
environment. They enable you to verify whether this collector is collecting data. For accurate analysis of this
information, send the collected data to the Information Warehouse and request analysis reports through the
Dashboard.
Troubleshooting Overview
As you use the VMware Capacity Planner collector, you might encounter problems or errors during the
discover, inventory, performance collection, and data synchronization processes. This is norma because of the
complex and variable nature of network configurations, system security issues, and more.
The information provided in this chapter will help you identify the cause of the problems you encounter.
Suggestions are given to assist you in correcting the problem, or working around it.
In addition to the information presented in this chapter, also consult the VMware Capacity Planner Release
Notes accompanying the product.
If you have a problem that is not discussed in this chapter or in the Release Notes, check the VMware
Knowledge Base at https://www.vmware.com/support/kb/. If none of these resources answer your problem,
report the problem to VMware Capacity Planner Technical Support.
The amount of information written to the log files is determined by the logging level defined in the Data
Manager. As the logging level increases, more detailed information about the activity of the collector is
recorded in the log file and displayed in the Message pane.
The log files are standard text files and can be opened with any text or word processor. When problems are
encountered, examine the log files first to determine whether an error message was reported, or if other
pertinent information was recorded. If you cannot find enough pertinent information in the log file to help
explain the problem, raise the logging level and repeat the procedure or other sequence of events that triggered
the problem.
The Log Level is set by the slider in the Options dialog box.
2 Move the slider to the setting you want and click OK.
As you move the slider, the level of logging (such as Follow Progress) appears. The further to the right the
setting is, the higher the logging level.
1 Follow Progress
3 Light Debugging
4 Detail Debugging
5 Trace Mode
As the logging level increases, so does the amount of detailed information that is recorded in the log file (and
also displayed in the Message pane of the Data Manager window).
CAUTION Prolonged logging at a high level can quickly fill your log file, consume available disk space, and
possibly affect performance of the collector. Limit use of detailed logging to troubleshooting a possible
problem with the collector.
Firewalls
Firewalls provide security for systems from intruders. They can also prevent access for legitimate
administrative users. If several systems are in a location behind a firewall, you might need to install the
collector and Data Manager on a host system within the firewall.
Unmanaged Systems
The network environment for a company might include some unmanaged systems that are connected to the
network, but not joined into any domain. To connect to an unmanaged system, the collector must use a local
administrative account for that system.
This scenario is not uncommon after a company merger or acquisition has occurred. A transition plan might
exist for defining the management of these new systems, but until the plan takes effect, many of these systems
are connected to the network, but not joined into any domain.
Error Codes
In addition to the information presented in this chapter, with the suggested solutions and workarounds, check
Appendix A, “Error Codes,” on page 127 for commonly-encountered error codes.
Discovery Troubleshooting
The discovery process relies on the network configuration at the target site, so if there are problems with the
discovery of a particular target system (or group of systems), the first step is to identify what is missing and
investigate the network configuration.
The collector uses the same APIs to populate the network neighborhood. If certain target domains or systems
are expected to be discovered, but are not found during the discovery process, go into the All Objects branch
in the Navigation pane of the collector host and verify that the expected domains and systems can be seen.
Possible workarounds to consider when target domains or systems are not discovered are:
If specific domains are missing, manually add the missing domains with the Data Manager. This often
results in discovery of systems in the added domains. If a file of missing domains and systems is available,
the Data Manager can import those domains and systems from the file.
For domains that are not providing complete server listings, ensure that the collector host has WINS
servers listed for those domains.
Create an LMHOSTS file with entries for missing domains and the corresponding primary domain
controllers.
If a list of target systems is available but the associated domains are unknown and it is too
time-consuming to identify the associated domains, you can create a dummy domain and the import the
target systems.
A target system within a Microsoft LAN Manager domain resolves with its short name, and Capacity
Planner can collect inventory and performance information without the proper domain name.
Bad Path
A bad path error is typically caused by the following problems:
Because these problems can manifest themselves in several ways, numerous possible solutions exist.
Name resolution is easy to detect and is usually straightforward to resolve. To correct this problem, make it
possible for the hostname indicated for a target system to be resolved to an IP address. To do this, correct a
DNS, WINS, or Active Directory name locator server, or edit the local LMHOST file to fix the problem. Use the
nslookup command to verify that the hostname is resolvable.
Change the LMHOST file, and not the HOST file, because Capacity Planner uses some APIs for LAN Manager
protocols.
Name resolution problems are not connectivity problems. Name resolution involves getting the correct
physical address for a system. Typical reasons for name resolution problems include:
The system no longer exists on the network, and has been gone long enough to be removed from Active
Directory, WINS, or browser synchronization.
If system names are imported, either a system name was misspelled or the system name is not resolvable.
The system was moved to another domain and because the collector host uses a fully qualified domain
name (FQDN) to contact the system, the name is incorrect.
If a short name is used for the path of a target system instead of the FQDN, the domain of the collector
host is appended to the end of that short name by default. If a bad path error is encountered, the collector
host might be in a domain that cannot resolve the remote system.
The best tools to verify the path to a target system are the ping and nbtstat commands. See “Useful
Troubleshooting Tools” on page 122.
Connection to the IP address of a target system might not be accomplished for the following reasons:
A system is not connected to the network and was not removed by the name resolution service. This is
common for Active Directory. Active Directory does not remove systems from the database when they go
offline.
The system is in a location that prevents connection either by router configuration or a firewall.
This problem can be difficult to diagnose. In many cases, one protocol or port is allowed, but another is
not. For example, you might be able to successfully ping a target system by its host name, yet you cannot
map a drive on that system.
Make sure that the protocols used by Capacity Planner are successful. Use the Test Connection menu
command in the Data Manager.
Unknown User
The unknown user error occurs when the user account that the collector is using is not known to a target
system. If the collector host is using a local account to connect to a remote system, the account probably does
not exist on that system. If the collector host is using a domain account:
The target system might belong to a different domain and might not have a trust relationship with the
domain to which the user account belongs.
The unknown user error might also occur because of a scenario that might be difficult to determine.
If a target system is a member of a domain that is discovered by using both LAN Manager and Active Directory
APIs, the credentials might conflict or be invalid during certain tasks when the collector accesses this system.
You must use the same account credentials to connect to the system through both of the discovered types: LAN
Manager and Active Directory.
This ensures that the collector can connect to the system for inventory and performance data collection because
the collector tries either path to reach the system. It always tries to find the quickest error-free path to a target
system, regardless of the domain to which it belongs.
Access Denied
The collector host can be denied access to a remote system if it is trying to connect with an invalid password.
Another consideration is whether the remote system is configured to allow remote access only for specific
users accounts or groups or only for specific domains or network paths.
You might receive an access denied message if there is no trust relationship for the domain. It is more likely
that an untrusted domain is indicated by an unknown user error.
Administrator Privileges
If the collector host is using an account that is not the local administrator account, verify that the account is
added to the Administrators group on the remote system.
Some environments default to a guest account if the provided user name and password are incorrect. This
gives the false security that the user account or password is correct and connection was successful. This can
lead to later problems with inventory and data collection because a guest does not have necessary privileges
to access the necessary data on the remote system.
to a local administrator group, create a trust relationship, or make sure that the local administrators group is
part of the domain administrators group.
Check whether the domain name was mistyped. If the domain name is incorrect, the remote system
cannot resolve any of the domain controllers for that domain.
Log in to the remote system with any account and then ping one of the domain controllers. This helps
determine whether the network path from the remote system to the domain is not accessible.
RPC Error
A remote procedure call (RPC) error indicates that a connection requested on a specific port was not
successful. RPC connection requests occur after name resolutions and IP connections.
The most common cause of an RPC error is that the RPC service is shut down.
1 Determine whether the host name is resolvable by executing the following command:
NBTSTAT –a hostname
2 If you can resolve the host name, try to connect to the system with the command:
net use \\hostname $IPC
If you cannot connect, one of the services—Remote Registry, WMI, or Perfmon—is disabled.
The net use command also confirms that the required ports are available for the collector to run inventory
and performance.
Network issues
Service failures
Authentication problems
Most issues that affect successful data collection fall into these categories.
You can also expect a certain order of failures. You can usually expect network issues to occur first, followed
by service issues, and then authentication issues.
Begin troubleshooting any data collection issue by checking the collector host first:
Network Issues
If the collector cannot connect to a remote system, it cannot determine whether a service is running or that a
connection account is valid. You must get the network working before trying anything else.
Routers
Network bandwidth
Firewall
Windows XP SP2
Name Services
If you are having trouble reaching a target system for data collection, make sure the name of the system can
be resolved to an IP address first. Sometimes this IP address can be the wrong one, because it is an old static
IP address. If the name service is providing the wrong IP address or no IP address, you never reach it. Firewalls
and routers can also be problematic because they close ports for security and scalability. In general, try to ping
the system, or use other network check utility programs like nbtstat and nmap.
Service Failures
A service failure can resemble a network issue because of the error message that is returned. If you are sure
that the network is available and other systems are on the same subnet that allow connection, a required
service is probably shut down.
Services run on a system and listen on a TCP or UDP port for a request. When a request occurs, the sevice
authenticates the request and responds appropriately. If you receive an access denied response, you confirmed
that the service is running and you can proceed to troubleshoot the connectivity problem as an authentication
failure.
Inventory uses WMI, Net APIs, Registry, Perfmon, and SSH. For each target system, inventory determines
what type of system it is: Windows, Linux, or UNIX. With this information, it tries the different protocols to
collect information. The tests are WMI, Registry, and SSH. If Registry is selected, Perfmon is used to determine
the amount of memory. Verify the list of required services before installation.
Authentication Failures
Authentication failures are usually indicated by error messages similar to access denied. The types of
authentication errors that occur are:
The collector host can usually discover these systems, but not access them to collect inventory or performance
data. To gain access, the security owners of the systems must be contacted to provide accounts that the
collector can use for accessing the necessary data from these systems.
To check the status of the Remote Registry service on the target system, choose Start > Control Panel >
Administrative Tools > Services.
In addition to checking the status of the service, check for an incorrect login, or a login that does not have
administrative rights on the target system. Verify that no firewall is blocking any traffic to or from the target
system.
1 Choose Start > Setting > Control Panel > Administrative Tools > Computer Management.
Icons for all user accounts on this computer appear in the right side of the Computer Management screen.
You should be able to collect performance data while the target machine has the administrator logged in.
Before you start collection, if the target machine is behind a firewall, turn off the firewall. Make sure that
services are started.
The expected result of the Inventory task is that WMI detects the CD-ROM, but classifies them as removable
drives. Inventory data collection is only interested in the hard drives on the target system because an
important performance statistic is the use of disk space on the system disks. Because the CD-ROMs are
reported as removable disks, they are not reported back in the inventory data that WMI collects.
When the Registry method of collection is used, the Registry reports that it detects these disks but doesn’t
know that they are CD-ROMs and are considered removable media. The same behavior is seen with Linux and
UNIX inventory collection, depending on the target system.
Perfmon Collection
Some performance data—such as, RAM or CPU information—must be collected by Perfmon, and Perfmon can
only be accessed by an administrator-level user. If the collector host is connecting to a system with a
non-administrator user, some expected performance data is not retrieved from that system.
2 Type gpedit.msc.
3 Choose Computer Configuration > Administrative Templates > Network > Network Connections >
Windows Firewall > Domain Profile.
6 Click OK.
You can change this default behavior by defining specific IP ranges or domains for the UNIX, Linux, and
Windows systems on your network.
1 From the Data Manager menu bar, select Admin > Options.
2 Click the Modules tab.
4 Click the Group/Domain Discovery tab and select DNS or click the Node Discovery tab and select IP
Addresses.
7 If you are defining domains, click the DNS Domains tab and add the domains you want the collector host
to work on.
8 If you are defining IP ranges, click the IP Subnets tab and add the subnets that you want the collector host
to scan.
Unless otherwise noted, detailed information about these commands and utilities is provided in the online
Help of your Windows system. These commands are described in relation to their use on the collector host.
ping
The ping command allows you to verify connections to a remote computer (or computers), confirming that
the remote system is online and responding.
Ping can be useful in determining whether your system is online, but a successful ping does not guarantee
that collection will be successful.
A failed ping can mean that the system or network does not allow Internet Control Message Protocol (ICMP)
traffic.
Example The following command tests the ability to connect to abcserver by sending packets to it and
waiting for a response from the remote system. The connection is made by resolving the host name of the
remote system.
ping abcserver
The results from this command indicate the degree of success of the connection attempt, and the round trip
time statistics.
nbtstat
The nbtstat command is a diagnostic command that helps determine how a system name or IP address is
resolved. Because it can display current connections using NetBios over TCP/IP (NBT), it is useful for
determining if Windows systems are online from a NETBIOS view.
Example The following command lists the name table for abcserver, resolving access to that remote system
by its host name:
nbtstat –a abcserver
net view
The net view command is one of a collection of network (net) commands. The net view command allows
you to display domains or computers, or shared resources, that can be accessed from the local system.
Example The following command determines which shared resources are available at abcserver.
net view \\abcserver
When used without parameters, net view displays a list of computers in the domain of the local system.
net use
The net use command is another Windows network command. Its typical use is to connect to or disconnect
from a remote computer or shared resource. It can also provide information about existing connections,
making it a useful tool to verify that the collector host can connect with this account.
A variation on the preceding command connects to abcserver for the purpose of remote administration, thus
ensuring that the abcuser account has administrator privileges on abcserver:
net use \\abcserver\admin$ /USER:abcuser
Perfmon
The Perfmon utility can help to verify that a remote system can be monitored. In addition, it can also add
performance counters from a remote system. However, Perfmon only works locally as a troubleshooting tool.
Because Remote Registry and File and Print services use the same ports as Perfmon, you can test access to a
target system with those tools. If Perfmon is run locally on a problem system, it allows you to verify that the
necessary services are running correctly and that the correct counters are installed for Performance data
collection.
tracert
Also called traceroute (or tracepath on some Linux systems), the tracert tool allows you to determine
the route IP packets use as they navigate an IP network. The tracert tool shows the route taken to reach a
particular destination on the network, and is thus a useful tool for troubleshooting the network.
It is helpful in determining whether WMI is working on the remote system. You can also verify your access
rights on a remote system by navigating through the console tree in different areas and looking for Access
Denied or Access is Denied errors.
To start the Computer Management utility, select Control Panel > Administrative Tools > Computer
Management.
To manage a remote system, right-click the top level of the navigation tree and select the remote system to
connect to.
Registry Editor
The Registry Editor is generally used to view and edit the Registry on your local system. However, it is also a
useful tool to test Windows Registry connectivity to a remote system.
Windows provides two forms of the Registry Editor: regedit.exe and regedt32.exe. Microsoft
recommends the use of regedit as the preferred utility for general browsing and searching of the Registry.
To start either form of the Registry Editor, select Start > Run and specify the utility name.
The Registry Editor window opens and displays the local Registry. To connect to the Registry of a remote
computer, select Connect Network Registry from the File menu.
CAUTION Be careful when you access a Windows Registry. If any key in the Registry is inadvertently (and
incorrectly) modified, operation of that Windows system can be severely affected.
PuTTY
PuTTY is a third-party shareware Win32 Telnet and SSH client. PuTTY allows you to open a terminal session to
remote systems that support SSH. This tool also includes certificate management tools that allow you to assign
your private keys. PuTTY and its related binaries are included in the third-party subdirectory of the Capacity
Planner installation folder.
Considerations for accessing UNIX and Linux systems, and the role of PuTTY in this process, are provided in:
To see limited data more quickly, change the Dashboard viewing interval from Weekly to Hourly. See the
VMware Capacity Planner Dashboard Users Guide for information about viewing data.
If you do not see any data from your collector host through the Dashboard, verify that the database ID for this
collector is properly registered in the Information Warehouse for the correct company.
Verify that the database ID for this collector is properly registered in the Information Warehouse for the
correct company. (See Chapter 5, “Setting Up the Collector,” on page 43.)
Make sure that your scheduled jobs are set up correctly on the collector host, and that the Scheduler is not
suspended. (See Chapter 5, “Setting Up the Collector,” on page 43.)
Make sure that you wait one full business day from setting up the collector host.
Check the Internet connection between the collector host and the Information Warehouse:
You must be able to send traffic through the HTTPS port (443) of your collector host.
Make sure that you can connect to an HTTPS site from the collector host.
Ping the Information Warehouse site from the collector host, to verify connectivity.
If a proxy server is on your network, from the Data Manager’s menu bar, select Admin > Options >
Connection and enter the proxy setting.
Select Use Manual Settings and enter the correct proxy setting.
Generally, database compaction is performed regularly without any problem. You might notice messages
appearing in the Message pane related to compaction, including an occasional message that database
compaction was not successful (the extent of this message depends on the logging level that is in effect).
A failed database compaction is not a problem. It usually occurs when the Data Manager or collector tries to
compact the database at the same moment that the other has the database locked for access during normal
operation.
Although it is not required, you can manually compact the database by selecting Admin > Database >
Compact Database.
4 Renames the temporary file (which is the compacted database) to the proper name of the current database.
NOTE The specific messages recording these steps can be seen in the Message Box by setting the Log Level to
level 3 (Light Debugging).
Purge Database
You can manually purge the database by selecting Admin > Purge Database. If you select this option, you are
prompted witha warning.
Timeouts
Requests to some target systems might timeout. To correct this, increase the timeout for one or more modules
at the task level. An example for the Inventory task follows.
1 From the Data Manager menu bar, select Admin > Options.
The Inventory task is listed on the Tasks tab with its Continue On setting and the task time limit, as
defined by the task timeout parameter defined for this task.
5 Select the Inventory task and click Modify.
Click Modify to change the necessary parameters for this task. Double-clicking the task name on the
Tasks tab does not start the necessary dialog box.
7 Set the task timeout and object timeout periods, then click OK.
Repeat this process for any other module that is timing out.
Error Codes A
This appendix documents common errors that you might encounter when running the VMware Capacity
Planner collector and Data Manager.
Common Errors
The majority of errors that are documented here occur during the inventory and performance data gathering
processes—it is those two tasks that require successful connection to target systems with permissions that
allow various utilities or system services to be run to obtain necessary information. For descriptions of these
errors, see “Data Collection Errors” on page 127.
The next most common period in which errors are encountered is that of data synchronization when the
collector transmits its collected data to the Information Warehouse. “Data Synchronization Errors” on
page 130 describes these errors.
The errors usually appear in the Message pane of the Data Manager window and are logged in the collector’s
log files.
A number of the messages have the same message text but different error codes. Compare the error code with
the message text when looking up an error in this appendix. The message you are looking for might be in
another section. For example, there are five possible Access Denied messages, each with a different error code.
The cause of these messages depends on what you are doing at the time.
If you encounter an error that is not documented in this chapter, contact VMware Capacity Planner Technical
Support as described in “Technical Support and Education Resources” on page 12.
-214702489 Problems with the connection Check the user account, as it is defined for the
Access is Denied account on the target system: collector, that no misspellings occurred in the
Account does not exist. account name or domain.
Note: Note the “is” in the
error message. The Password is invalid. Retype the password in the collector to make sure
number and the message User needs to change that it is correct.
are different than if the password on first login. Note: The connection account might have been
user is valid, but does not defined in the setup options as a global connection
Account is disabled.
have enough rights (error for the collector, or as a custom connection account
User domain cannot be
-2147217405: Access for the specific target system.
Denied). reached for authentication.
Log in to the target system with the account to
verify that everything is working.
Check the user account in the domain or target
system to verify that it exists, is not disabled, or
does not require a password change on first login.
Check that the target system is really in a domain
and not in a workgroup with the same name.
If the user account is not in the same domain as the
target system, ensure that the proper domain trusts
exist.
2147024882 This error happens on systems
Inventory error running Windows NT when the
Network Redirector runs out of
buffer space when processing a
command and the IRPstack is too
small for the command to
complete.
2147217400 Problems with a WMI driver Perform inventory collection using the Registry
Invalid parameter, WMI and not WMI.
error
-2147217405 Problems with the connection Add the connection account to the local
Access Denied account on the target system: administrators group on the target system.
Note: Compare to error User does not have enough If the user is in a domain administrators group, add
-214702489: Access is rights. the domain administrators group to the local
Denied) User not in the local administrators group on the target system.
administrator group on the
target system.
Inherited rights for the user
are not sufficient.
2147217407 Problems with a WMI driver Perform inventory collection using the Registry
Failure to collect and not WMI.
inventory data
-2147221164 Target system is one of the The collector handles this scenario and switches to
Class not registered following: Registry when this happens (with default settings).
========== NT 4.0 with no WMI This is normal behavior for NT 4.0 and there is nothing
Windows 2000 or higher to do since WMI is not available.
-2147217394
with WMI stopped For Windows 2000 or later:
Error description not
found If WMI is shut down for security reasons, don’t do
anything.
If WMI is shut down for testing reasons, restart the
service.
-2147217392 The class of items does not Usually, there is nothing to do in this situation. The
Invalid Class exist. collector continues to collect information from objects
========= Older hardware that it finds. In some cases, the collector intentionally
tests for the existence of an object before it goes on.
438 Specific hardware drivers do
Object doesn’t support not populate WMI
this property or method Possible language issue with
class names
-2147023174 Target system is offline Ping the target system and fix any network or service
The RPC Server is The remote procedure call issue.
unavailable (RPC) service is stopped on
collector host or on target
system)
Required ports are blocked
2147467259 The database has been Ensure that the Microsoft Jet Database engine can be
Not collecting inventory placed in a state that opened and logged.
prevents it from being
opened or logged
2147481648 May need another collector If systems that are failing appear to be mostly in one
Unable to run on the domain that has domain, add another collector for that domain. Add an
performance in some systems failing account that gives you access to the systems in the new
servers domain.
5 Usually occurs during inventory data collection: Make sure the connection
Access denied This is probably a registry access failure: The user has the rights to account has
connect, but does not have enough rights to read the necessary administrative rights on
keys. the target system.
An attempt was made to log in, Can occur during either inventory or performance data collection: Check the Services for the
but the network login service Net logon service is not running. system and make sure
was not started that the Net login services
are running.
Command = ‘[command name]’ This error occurs during data collection on UNIX or Linux Install the required
returned Error = [error that machines (or when data collected from these machines using the tools/commands on the
occurred] collector scripts is imported by data synchronization). The error target machine indicated
occurs when scripts used by the collector cannot find tools or after HOSTNAME=. The
commands the script expects to find on the target machine. The TOOLS block text
An example follows:
error message shown on the left appears in the message box on the provides information to
Command = ‘isdev’ returned collector host. help you troubleshoot
Error = not found in
The result is that some data cannot be collected, and a TOOLS block and fix the problem.
/bin:/usr/bin:/usr/sbin:/sbin:/usr
appears in the collected data files. (Data files are typically found in
/contrib/bin
C:\Program Files\VMware\VMware Capacity Planner\Temp
until the data is imported into the collector. If the Archive option
is selected, the files are copied to C:\Program Files\VMware
\VMware Capacity Planner\archive).
An example of a TOOLS block follows:
<TOOLS>
HOSTNAME=cp-linux1
FILETYPE=Log
Date=Tue Oct 23 16:20:13 PST 2007
TIMESTAMP=20070206162013
CMD=showmount
ERR=mount clntudp_create: RPC: Program not registered
</TOOLS>
Data synchronization errors can occur when the collector attempts to transmit data files to the Information
Warehouse. The error messages are listed alphabetically in Table A-3.
5 A file transfer was attempted and did not Notify VMware of the file causing the error so
Access denied successfully complete: that the file can be deleted from the upload
VMware does not allow files to be directory.
overwritten and the collector is trying Check the rights to the archive directory to make
to resend a file with the same name. sure the collector can write to it.
After the file transfer, the collector is
trying to move the file to the archive
directory, but the directory is
read-only.
Somehow the file being sent is marked
as read-only and the collector cannot
delete it.
Can’t send data Invalid proxy settings Verify that the Information Warehouse site
to VMware (https://optimize.vmware.com) is available and
you can log in to the Dashboard.
Check proxy and port settings in the Data
Synchronization job (In the Data Manager, select
Admin > Options > Connection).
You might need to set the Service User account for the
collector service to an account that has rights to
browse the Web outside the company.
Miscellaneous Errors
The error messages are listed alphabetically in Table A-4.
713 Capacity Planner was installed as a When you receive the error message dialog box that
Class not different user than what you are running. shows the .dll name, run:
registered Might cause the Data Manager to fail. regsvr32 full.dllpath
Some dlls need to be registered for each For example:
user. regsvr32 c:\winnt\system32\msstdfmt.dll
In the Data MDAC is too old Upgrade to the latest version of MDAC from
Manager, the http://msdn.microsoft.com/data/ref/mdac/downloads/
navigation tree This problem typically does not exist in Capacity
does not expand Planner version 2.1.0 or later because the installer
to show the checks for MDAC and makes sures that it is up-to-date.
objects If it is not, the installer automatically installs the
underneath required version.
1726, 1789 Usually, an error message and code are converted and available to view on the Properties tab of the
Failed: unknown. Details area for the target system on which the error occurred.
If it’s not showing up there, it is probably an error code that Capacity Planner has not encountered
before. If this occurs, contact VMware Technical Support as described in “Technical Support and
Education Resources” on page 12.
To address this problem, Technical Support needs to evaluate a copy of your Capacity Planner log
file.
Chapter 4, “Installing the Collector,” on page 33 provides the instructions for installing the collector and Data
Manager components of VMware Capacity Planner. It also discusses the considerations that you must make
if you are installing Capacity Planner on a system with a previous Capacity Planner installation. (See
“Upgrading a Collector Installation” on page 36.)
Deletes the service from the system (and the list of system services).
4 To confirm that the collector service is uninstalled, select Admin > Collector menu.
6 Proceed with removing the remaining components of the Capacity Planner application.
1 In the Windows Control Panel, double-click the Change or Remove Programs icon.
2 Scroll through the list of currently installed programs to find VMware Capacity Planner.
The InstallShield Wizard displays the message that it is preparing itself, then prompts you to confirm that
you want to uninstall the Capacity Planner application.
If you did not uninstall the collector service from the Data Manager before beginning this removal
process, the collector service has to be stopped before you remove it.
When the Capacity Planner application is removed, a window appears that indicates that the Installation
wizard completed its tasks:.
6 Click Finish.
1 Start a Windows Explorer window and navigate to the folder in which the Capacity Planner application
was installed.
2 Select all files in the installation folder and delete those files.
4 (Optional) Remove the empty installation folder and all empty subfolders.
Inventory Data C
This appendix provides a list of the inventory data that you can collect from the various operating systems
(Linux and UNIX, Registry for MS Windows NT 4.0, and WMI for MS Windows 2000/XP) and tells you if you
can view the data in the Data Manager or the Capacity Planner Dashboard.
Performance Counters
This section lists many of the performance counters that the collector can gather. These performance counters
appear in the Dashboard performance tables.
The list of performance counters in this section is not an exhaustive list of all counters the Dashboard might
display. The performance counters are expected to change over time. Much of this information is from
third-party vendors and VMware is not responsible for its accuracy. It is included here as a convenience.
For information about how performance counters are used by VMware Capacity Planner, see Chapter 9,
“Collecting Performance Data,” on page 95. The performance counters are grouped alphabetically by the
following performance objects:
Active Server Pages Memory Allocated Amount of memory allocated to this process.
Requests Executing
Requests Queued Number of requests waiting for service from the queue.
Sessions Total
Server List Requests/sec Rate at which this workstation processed requests to retrieve
a list of browser servers.
Copy Read Hits% Percentage of cache copy read requests that hit the cache (that
is, they did not require a disk read to provide access to the
page in the cache).
A copy read is a file read operation that is satisfied by a
memory copy from a page in the cache to the applications
buffer.
The LAN Redirector uses this method for retrieving
information from the cache as does the LAN Server for small
transfers. This is a method used by the disk file systems as
well.
Data Maps/sec
Data Flushes/sec Rate at which the file system cache has flushed its contents to
disk as the result of a request to flush or to satisfy a request to
request a write-through write a write-through file more than
one page on each flush operation.
Data Flush Pages/sec Number of pages per second the file system cache has flushed
to disk as a result of a request to flush or to satisfy a to request
a write-through file more than one page on each flush
operation.
Data Map Hits % Percentage of data maps in the file system cache that can be
resolved without having to retrieve a page from the disk
because the page was already in physical memory.
Fast Reads/sec
Lazy Write Pages/sec Rate at which the Lazy Writer thread has written to disk. Lazy
Writing is the process of updating the disk after the page was
changed in memory so that the application that changed the
file does not have to wait for the disk write to be complete
before proceeding. More than one page can be transferred on
a single disk write operation.
If the counter value is increasing over time, this can indicate
that memory is becoming low.
Cache Lazy Write Pages are a subset of Data Flush Pages.
MDL Reads/sec
MDL Read Hits % Percentage of Memory Descriptor List (MDL) Read requests
to the file system cache that reach the cache (that is, the file
system requests did not require disk accesses to provide
memory access to the pages in the cache).
Pin Reads/sec
Pin Read Hits % Percentage of pin read requests that reach the file system
cache (that is, did not require a disk read to provide access to
the page in the file system cache).
While it is pinned, a page’s physical address in the file system
cache is not altered.
The LAN Redirector uses this method for retrieving data
from the cache as does the LAN Server for small transfers.
This is usually the method used by the disk file systems as
well.
Read Aheads/sec Frequency of reads from the file system cache in which the
Cache detects sequential access to a file.
The read aheads permit the data to be transferred in larger
blocks than the blocks that the applicationrequests, reducing
the overhead for each access.
Citrix IMA Networking Bytes Received/sec Number of inbound bytes per second.
DataStore bytes read/sec Number of bytes of data store data read per second.
DataStore bytes Number of bytes of data store data written per second.
written/sec
DataStore reads/sec Number of times data was read from the data store per
second.
DataStore writes/sec Number of times data was written to the data store per
second.
DynamicStore bytes Number of bytes of dynamic store data read per second.
read/sec
DynamicStore bytes Number of bytes of dynamic store data written per second.
written/sec
DynamicStore reads/sec Number of times data was read from the dynamic store per
second.
DynamicStore Number of times data was written to the dynamic store per
writes/sec second.
LocalHostCache bytes Number of bytes of IMA local host cache data read per
read/sec second.
LocalHostCache bytes Number of bytes of IMA local host cache data written per
written/sec second.
LocalHostCache Number of times data was read from the IMA local host cache
reads/sec per second.
LocalHostCache Number of times data was written to the IMA local host cache
writes/sec per second.
Zone Elections Number of zone elections that occurred. This value starts at
zero each time the IMA service starts and is incremented each
time a zone election takes place.
Zone Elections Won Number of times the server won a zone election.
Bytes In/Sec
Btyes Out/Sec
Cache Pops/Sec
DB Hits/sec
Page hits/Sec
Queued Requests
Running Requests
Bytes In/Sec
Btyes Out/Sec
DB Hits/sec
Page hits/Sec
Queued Requests
Running Requests
Database Cache % Hit Percentage of Directory pages satisfied from the Directory
Cache
DHCP Server Acks/sec Rate at which DHCP acknowledgements were sent by the
DHCP server.
Active Queue Length Number of packets in the processing queue of the DHCP
server.
Conflict Check Queue Number of packets waiting in the DHCP server queue
Length because of conflict detection.
Duplicates Dropped/sec Rate at which the DHCP server received duplicate packets.
Packets Expired/sec Rate at which packets expire in the DHCP server message
queue.
Packets Received/sec Rate at which packets are received by the DHCP server.
Split IO/sec
% Free Space
% Tiime
% Write Time
Disk Reads/sec
Disk Transfers/sec
Disk Writes/sec
Free Megabytes
% Disk Read Time Percentage of elapsed time that the selected disk drive was
busy servicing read requests.
% Free Space Percentage of total usable space on the selected logical disk
drive that was free.
Avg. Disk Queue Length Average number of both read and write requests that were
queued for the selected disk during the sample interval.
Current Disk Queue The number of requests outstanding on the disk at the time
Length the performance is collected. It also includes requests in
service at the time of the collection. This is an instantaneous
snapshot, not an average over the time interval. Multispindle
disk devices can have multiple requests that are active at one
time, but other concurrent requests are awaiting service. This
counter might reflect a transitory high or low queue length,
but if there is a sustained load on the disk drive, it is likely
that the counter is consistently high. Requests experience
delays proportional to the length of this queue minus the
number of spindles on the disks. For good performance, this
difference averages less than two.
Memory (continued) Available Bytes Amount of physical memory available to processes running
on the computer in bytes, calculated by summing space on
the Zero, Free, and Stand by memory lists.
Free memory is ready for use.
Zeroed memory are pages of memory filled with zeros to
prevent later processes from seeing data that a previous
process uses.
Standby memory is memory that is removed from a process
working set (its physical memory) on the way to disk, but that
is still available to be recalled.
This counter displays the last observed value only it is not an
average.
Cache Bytes Sum of the System Cache Resident Bytes, System Driver
Resident Bytes, System Code Resident Bytes, and Pool Paged
Resident Bytes counters.
This counter displays the last observed value only; it is not an
average.
Cache Bytes Peak Maximum value of Cache Bytes since the system was last
restarted.
This value might be larger than the current size of the cache.
Cache Bytes is the sum of the System Cache Resident Bytes,
System Driver Resident Bytes, System Code Resident Bytes,
and Pool Paged Resident Bytes counters. This counter
displays the last observed value only; it is not an average.
Cache Faults/sec Rate at which faults occur when a page sought in the file
system cache is not found there and must be retrieved from
elsewhere in memory (a soft fault) or from disk (a hard fault).
The file system cache is an area of physical memory that
stores recently used pages of data for applications.
Cache activity is a reliable indicator of most application I/O
operations.
This counter counts the number of faults without regard for
the number of pages faulted in each operation.
Demand Zero Faults/sec Rate of page faults that require a zeroed page to satisfy the
fault.
Zeroed pages are pages that is emptied of previously stored
data and filled with zeros; they are a security feature of
Windows NT. They prevent processes from seeing data
stored by earlier processes that used the memory space.
Windows NT maintains a list of zeroed pages to accelerate
this process.
This counter counts numbers of faults without regard to the
numbers of pages retrieved to satisfy the fault.
It displays the difference between the values observed in the
last two samples divided by the duration of the sample
interval.
Free System Page Table Number of page table entries that they system is not currently
Entries using. This counter displays the last observed value only; it is
not an average.
Memory (continued) Page Faults/sec Overall rate at which the processor handles faulted pages
measured in numbers of pages faulted per second.
A page fault occurs when a process requires code or data that
is not in its working set (its space in physical memory).
This counter:
Includes hard faults (those that require disk access) and
soft faults (where the faulted page is found elsewhere in
physical memory).
Displays the difference between the values observed in
the last two samples divided by the duration of the
sample interval.
Most processors can handle large numbers of soft faults
without consequence. However, hard faults can cause
significant delays.
Page Faults/sec Rate at which pages are read from or written to disk to
resolve hard page faults.
Hard page faults occur when a process requires code or data
that is not in its working set or elsewhere in physical memory
and must be retrieved from disk.
This counter is:
A primary indicator of the kinds of faults that cause
system-wide delays.
The sum of Memory: Pages Input/sec and Memory: Pages
Output/sec.
Counted in numbers of pages so it can be compared to
other counts of pages such as Memory: Page Faults/sec
without conversion.
Includes pages retrieved to satisfy faults in the file system
cache usually requested by application’s non-cached
mapped memory files.
Displays the difference between the values observed in
the last two samples divided by the duration of the
sample interval.
Pages Output/sec
Pool Nonpaged Allocs Number of calls to allocate space in the nonpaged pool.
The nonpaged pool is an area of system memory area for
objects that cannot be written to disk and must remain in
physical memory as long as they are allocated.
It is measured in numbers of calls to allocate space regardless
of the amount of space allocated in each call.
This counter displays the last observed value only; it is not an
average.
Pool Nonpaged Bytes Number of bytes in the nonpaged pool an area of system
memory (physical memory that the operating system uses)
for objects that cannot be written to disk but must remain in
physical memory as long as they are allocated.
This counter is calculated differently than Process: Pool
Nonpaged Bytes, so it might not equal
Process: Pool Nonpaged Bytes:_Total.
This counter displays the last observed value only; it is not an
average.
Memory (continued) Pool Paged Allocs Number of calls to allocate space in the paged pool.
The paged pool is an area of system memory (physical
memory that the operating system uses) for objects that can
be written to disk when they are not being used.
It is measured in numbers of calls to allocate space regardless
of the amount of space allocated in each call.
This counter displays the last observed value only; it is not an
average.
Pool Paged Bytes Number of bytes in the paged pool. The paged pool is an area
of system memory (physical memory used by the operating
system) for objects that can be written to disk when they are
not being used.
Memory: Pool Paged Bytes is calculated differently than
Process: Pool Paged Bytes so it might not equal
Process: Pool Paged Bytes: _Total.
This counter displays the last observed value only; it is not an
average.
System Cache Resident Size, in bytes, of the pageable operating system code in the
Bytes file system cache; this value includes only current physical
pages and does not include any virtual memory pages not
currently resident, so it might be smaller than the actual
amount of virtual memory in use by the file system cache.
This value is a component of Memory: System Code Resident
Bytes.
This counter displays the last observed value only; it is not an
average.
System Code Total Bytes Number of bytes of pageable operating system code
currently in virtual memory.
This counter is a measure of the amount of physical memory
that the operating systemuses that can be written to disk
when not in use. It is calculated by summing the bytes in
Ntoskrnl.exe, Hal.dll, the boot drivers, and file systems
loaded by Ntldr/osloader.
This counter does not include code that must remain in
physical memory and cannot be written to disk.
This counter displays the last observed value only; it is not an
average.
Memory (continued) System Driver Resident Number of bytes of pageable physical memory being used by
Bytes device drivers; it is the working set (physical memory area) of
the drivers.
This value is a component of Memory: System Driver Total
Bytes which also includes driver memory that has been
written to disk.
Neither System Driver Resident Bytes nor System Driver
Total Bytes includes memory that cannot be written to disk.
System Driver Total Number of bytes of pageable virtual memory currently being
Bytes used by device drivers. (Pageable memory can be written to
disk when it is not being used).
It includes physical memory (Memory: System Driver
Resident Bytes) and code and data paged to disk. It is a
component of Memory: System Code Total Bytes.
This counter displays the last observed value only; it is not an
average.
Transition Faults/sec Rate at which page faults are resolved by recovering pages
that were being used by another process sharing the same
page, or were on the modified page list or the standby list, or
were being written to disk at the time of the page fault. The
pages were recovered without additional disk activity.
Transition faults are counted in numbers of faults without
regard for the number of pages faulted in each operation.
This counter displays the difference between the values
observed in the last two samples divided by the duration of
the sample interval.
MSExchange Internet Bytes Received/sec Number of bytes the server received from the network.
Protocol
Bytes Sent/sec Rate at which bytes are sent on the interface, including
framing characters.
Incoming Queue Length Indicates the length of the Incoming packet queue.
Outgoing Queue Length Indicates the length of the output packet queue.
Incoming Inter-Site
Replication Updates/sec
MSExchangeIS Active Connection Indicates the number of connections to the Exchange store
Count that have shown activity in the last 10 minutes.
Active User Count Indicates the number of user connections that have shown
activity in the last 10 minutes.
MSExchangeIS Private Receive Queue Size Number of messages in the mailbox store's receive queue.
Send Queue Size Number of messages in the mailbox store's send queue.
MSExchangeIS Public Message Recipients Rate at which recipients are receiving messages.
Delivered/min
Receive Queue Size Number of messages in the public store's receive queue.
Send Queue Size Number of messages in the public store's send queue.
Total Size of Size of the recoverable items for the specified objects.
Recoverable Items
MSExchangeMTA LAN Receive Bytes/sec Rate at which bytes are received over a LAN from MTAs.
LAN Transmit Bytes/sec Rate at which bytes are transmitted over a LAN to MTAs.
Work Queue Length Queue length for the whole of the MTA. It covers both
inbound and outbound messages for the Information Store,
the Directory and any connectors that route through the
MTA.
XAPI Receive Bytes/sec Rate at which bytes are received over a XAPI connection.
XAPI Transmit Bytes/sec Rate at which bytes are transmitted over a XAPI connection.
Bytes Sent/sec
Objects Events
Mutexes
Sections
Semaphores
Physical Disk % Disk Read Time Percentage of elapsed time that the selected disk drive is busy
servicing read requests.
% Disk Time Percentage of elapsed time that the selected disk drive was
busy servicing read or write requests.
% Disk Write Time Percentage of elapsed time that the selected disk drive was
busy servicing write requests.
% Idle Time Percentage of time during the sample interval that the disk
was idle.
Avg. Disk Bytes/Read Average number of bytes transferred from the disk during
read operations.
Avg. Disk Queue Length Average number of read and write requests that were queued
for the selected disk during the sample interval.
Current Disk Queue Number of requests outstanding on the disk at the time the
Length performance data is collected.
This counter includes requests in service at the time of the
snapshot.
This is an instantaneous length, not an average over the time
interval. Multispindle disk devices can have multiple
requests active at one time but other concurrent requests are
awaiting service.
This counter might reflect a transitory high or low queue
length, but if a sustained load is on the disk drive, the counter
probably is consistently high.
Requests are experiencing delays proportional to the length
of this queue minus the number of spindles on the disks.
This difference averaged less than 2 for good performance.
Disk Bytes/sec Rate at which bytes are transferred to and from the disk
during write and read operations.
Disk Read Bytes/sec Rate at which bytes are transferred from the disk during read
operations
Disk Write Bytes/sec Rate at which bytes are transferred to the disk during write
operations.
Current Requests
Completed
Current Requests
Received
Total Requests
Completed
Total Requests Received
Print Queue Bytes Printed/sec Number of bytes per second that are printed from a print
queue.
Total Jobs Printed Total number of jobs printed from a print queue since the last
restart.
Total Pages Printed Total number of pages printed through GDI on a print queue
since the last restart.
Elapsed Time Total elapsed time (in seconds) this process has been running.
Page File Bytes Current number of bytes used by this process in the paging
files. If no paging file is present, this counter reflects the
current amount of virtual memory that the process has
reserved for use in physical memory.
Paging files are used to store pages of memory that the
process usesthat are not contained in other files.
All processes share paging files and a lack of space in paging
files can prevent other processes from allocating memory.
Page File Bytes Peak Maximum number of bytes this process used in the paging
files.
Private Bytes Current number of bytes this process has allocated that
cannot be shared with other processes.
Virtual Bytes Current size, in bytes, of the virtual address space for the
process.
Use of virtual address space does not necessarily imply
corresponding use of either disk or main memory pages.
Virtual space is finite; by using too much, the process can
limit its ability to load libraries.
Virtual Bytes Peak Maximum number of bytes of virtual address space the
process has used at any one time.
Use of virtual address space does not necessarily imply
corresponding use of either disk or main memory pages.
Virtual space is finite; by using too much, the process can
limit its ability to load libraries.
Processor % DPC Time Percentage of time that the processor spent receiving and
servicing deferred procedure calls (DPCs) during the sample
interval.
DPCs are interrupts that run at a lower priority than standard
interrupts.
% DPC Time is a component of % Privileged Time because
DPCs are executed in privileged mode. They are counted
separately and are not a component of the interrupt counters.
This counter displays the average busy time as a percentage
of the sample time.
% Interrupt Time The percentage of time the processor spent receiving and
servicing hardware interrupts during the sample interval.
This value is an indirect indicator of the activity of devices
that generate interrupts such as the system clock, the mouse,
disk drivers, data communication lines, network interface
cards and other peripheral devices.
These devices normally interrupt the processor when they
have completed a task or require attention. Normal thread
execution is suspended during interrupts.
Most system clocks interrupt the processor every 10 ms
creating a background of interrupt activity.
This counter displays the average busy time as a percentage
of the sample time.
% Interrupt Time
% Priveleged Time
% Processor Time
% User Time
APC Bypasses/sec Rate at which Kernel APC interrupts were avoided. APC
Bypasses/sec is the rate at which kernel APC interrupts were
short-circuited.
This counter displays the difference between the values
observed in the last two samples divided by the duration of
the sample interval.
DPC Rate Rate at which DPCs are added to the processor's DPC queue
between the timer ticks of the processor clock.
DPCs are interrupts that run at a lower priority than standard
interrupts. Each processor has its own DPC queue.
This counter:
Measures the rate at which DPCs are added to the queue not
the number of DPCs in the queue.
Displays the last observed value only; it is not an average.
DPCs Queued/sec Overall rate at which DPCs are added to the processor's DPC
queue.
DPCs are interrupts that run at a lower priority than standard
interrupts. Each processor has its own DPC queue.
This counter:
Measures the rate at which DPCs are added to the queue not
the number of DPCs in the queue.
Displays the difference between the values observed in the
last two samples divided by the duration of the sample
interval.
Redirector Bytes Total/sec Rate at which the Redirector is processing data bytes.
This includes all application and file data in addition to
protocol information such as packet headers.
Read Bytes Cache/sec Rate at which applications are accessing the file system cache
by using the Redirector.
Some of these data requests are satisfied by retrieving the
data from the cache.
Requests that miss the Cache cause a page fault (see Read
Bytes Paging/sec).
Read Bytes Paging/sec Rate at which the Redirector is attempting to read bytes in
response to page faults.
Page faults are caused by loading modules (such as programs
and libraries) by a miss in the Cache (see Read Bytes
Cache/sec) or by files directly mapped into the address space
of applications (a high-performance feature of Windows NT).
Write Bytes Cache/sec Rate at which applications on your computer are writing to
the file system cache by using the Redirector.
The data might not leave your computer immediately; it can
be retained in the cache for further modification before being
written to the network. This saves network traffic.
Each write of a byte into the cache is counted here.
Write Bytes Paging/sec Rate at which the Redirector is attempting to write bytes
changed in the pages being used by applications.
The program data changed by modules (such as programs
and libraries) that were loaded over the network are paged
out when no longer needed. Other output pages come from
the file system cache (see Write Bytes Cache/sec).
Server Bytes Received/sec Number of bytes that the server has sent to and received from
the network.
This value provides an overall indication of how busy the
server is.
Bytes Total/sec Number of bytes that the server has sent to and received from
the network.
This value provides an overall indication of how busy the
server is.
Bytes Transmitted/sec Number of bytes that the server has sent on the network.
This value provides an overall indication of how busy the
server is.
Pool Nonpaged Failures Number of times that allocations from nonpaged pool have
failed.
Indicates that the computers physical memory is too small.
Pool Paged Failures Number of times that allocations from paged pool have
failed.
Indicates that the computers physical memory or paging file
are too small.
Sessions Errored Out Number of sessions that were closed because unexpected
error conditions or sessions that reached the auto-disconnect
timeout and were disconnected normally.
Server (continued) Sessions Timed Out Number of sessions that were closed because idle time
exceeding the AutoDisconnect parameter for the server.
Shows whether the AutoDisconnect setting is helping to
conserve resources.
Server Work Queue Queue Length Current length of the server work queue for this CPU.
A sustained queue length > 4 might indicate processor
congestion.
This is an instantaneous count not an average over time.
Page Splits/sec Number of page splits occurring as the result of index pages
overflowing.
SQL Server:Buffer Manager Buffer Cache Hit Ratio Percentage of pages found in the buffer pool, thus saving a
read from disk.
Log File(s) Size (KB) Cumulative size of all the data files in the database.
SQL Server:Locks Average wait time (ms) The average amount of wait time (milliseconds) for each lock
request that resulted in a wait.
SQL Server:Memory Connection Memory Total amount of dynamic memory the server is using for
Manager (KB) maintaining connections
Lock Memory (KB) Total amount of dynamic memory the server is using for
locks.
Optimize Memory
SQL Cache memory Total amount of dynamic memory the server is using for the
(KB) dynamic SQL cache.
Target Server Memory Total amount of dynamic memory the server is willing to
(KB) consume.
Total Server Memory Total amount of dynamic memory the server is currently
(KB) consuming
Uploaded Changes/sec
System % Total Interrupt Time
Active Session
Context Switches/sec Combined rate at which all processors on the computer are
switched from one thread to another.
Context switches occur when a running thread voluntarily
relinquishes the processor for one of the following reasons:
It is preempted by a higher-priority ready thread.
It switches between user-mode and privileged (kernel) mode
to use an Executive or subsystem service.
This counter is the sum of Thread: Context Switches/sec for
all threads running on all processors in the computer and is
measured in numbers of switches.
Context switch counters are on the System and Thread
objects.
This counter displays the difference between the values
observed in the last two samples divided by the duration of
the sample interval.
Inactive Session
Processes Number of processes in the computer at the time of data
collection.
This is an instantaneous count, not an average over the time
interval. Each process represents the running of a program.
Processor Queue Length Number of threads in the processor queue.
Processor time has a single queue even on computers with
multiple processors.
Unlike the disk counters, this counter counts ready threads
only, not threads that are running.
A sustained processor queue of less than 10 threads per
processor is normally acceptable.
This counter displays the last observed value only; it is not an
average.
System Up Time Elapsed time (in seconds) that the computer was running
since it was last started.
This counter displays the difference between the start time
and the current time.
Web Service Bytes Total/sec Sum of Bytes Sent/sec and Bytes Received/sec.
This is the total rate of bytes transferred by the Web service.
Get Requests/sec Rate at which the Web service receives Get Requests.
Lock Requests/sec Rate at which the Web service receives Lock Requests.
Logon Attempts/sec Rate at which the Web service receives Login Attempts.
Post Requests/sec Rate at which the Web service receives Post Requests.
Put Requests/sec Rate at which the Web service receives Put Requests.
Search Requests/sec Rate at which the Web service receives Search Requests.
Unlock Requests/sec Rate at which the Web service receives Unlock Requests.
Logical Disk:Derived Writes/Split Number of writes that occur before a split I/O.
MSExchange:Derived Avg WBytes RPC Client Average number of write bytes through RPC per client.
Physical Disk:Derived Write Bytes/Split Number of write bytes before a split I/O occurs.
Read Aheads/Transaction Number of read aheads that occur for each SQL transaction.
“Commands Used for Linux and UNIX Inventory Collection” on page 163
The list of commands, utilities, and tools in this section is not an exhaustive list of all that might be in use. This
information is included as a convenience to assist you in working with Capacity Planner.
Inventory Utilities
This section lists many of the utilities the collector uses to collect UNIX and Linux inventory data.
Solaris pkginfo
Linux /proc/cupinfo
Solaris prtconf
Solaris prtdiag
Solaris ifconfig
Performance Utilities
This section lists many of the utilities the collector uses to collect UNIX and Linux performance data.
Linux, VMware
Solaris
Solaris
Linux df
Solaris
VMware
Solaris swap
Solaris iostat
Solaris netstat
Process HP-UX ps
Linux, VMware
Solaris
basename Yes No No
printf Yes No No
egrep Yes No No
cat Yes No No
sleep Yes No No
cut Yes No No
bc Yes No No
Monitoring Tools
This section lists tools that are used by the collector for collection of Linux performance data.
vmstat
Table E-4. vmstat on Linux
Used for
Class Metric Instance Consolidation
System Processor Queue Length N/A Yes
proc/swaps
Table E-5. proc/swaps on Linux
Used for
Class Metric Instance Consolidation
iostat
Table E-6. iostat on Linux
Used for
Class Metric Instance Consolidation
netstat
Table E-7. netstat on Linux
Used for
Class Metric Instance Consolidation
Import Classes
This section lists import classes that the collector uses to collect UNIX and Linux performance data.
SYSTEM Describes how the system is configured. Most of the information can be considered
environmental.
MOTHERBOARD Describes the chassis and motherboard that the system is built around.
GROUP Describes the groups of which the system is a member.
RAM Describes the physical memory installed in the system or the virtual memory allocated to a
virtual system.
FILESYS Describes the found file systems currently defined on the system.
DISKINFO Describes the physical drives or RAID arrays installed in the system.
DAEMON Describes the defined init.d, rec.d, or Microsoft Windows services on the system.
NETWORK Describes the physical and virtual network interfaces on the system.
EXPORTS Describes the exported or shared directories on the system.
PERF Contains detailed performance data that was collected during a single collection run.
UNIQID Unique ID string A unique ID that identifies the system. If provided will be
used to identify the system instead of the host name.
If you do not provide a unique ID, you might overwrite an
existing host name. If you have no unique ID, use an FQDN
host name.
RUNBY User information for the user who ran the script to generate
the output. Useful for debugging. In UNIX, use the ID
command.
ISRV_Type 2048, 32768 Number that tells VMware what kind of system is being
managed.
DAPP_Type App The table that contains stores application information. Also
used for operating system information.
ISA_Identity Identity string Used to uniquely identify the application.
ISA_Estimated Size Bytes number Size in bytes of the installed application. Usually the size at
installation.
DAPP_Producer Company string Name of the manufacturer that developed or packaged the
application.
DAPP_MajorVersion Major version number Major product version number, which is usually the first
number of the version string.
DAPP_MinorVersion Minor version number Minor product version number, which is usually the second
number of the version string.
DAPP_PatchLevel Patch level or Patch level or maintenance number for the product, which is
maintenance number usually the remainder of the version string.
ISA_InstallLoc Path string Filesystem path where the operating system is installed.
DAPP_Producer Company string Name of the manufacturer that developed or packaged the
operating system.
DAPP_MinorVersion Minor version number Minor operating system version number, which is usually
the second number of the version string.
DAPP_PatchLevel Patch level or Patch level or maintenance number for the operating
maintenance number system, which is usually the remainder of the version
string.
ISRV_DisplayName Host name string Alternate display name for the system.
ISRV_HostName Host name string Used to visually identify the system. Derived from the actual
host name.
ISRV_InstallLocale Install locale number Locale that was chosen during the installation. Applies more
to Window than UNIX. US English is 1033.
ISRV_CurrentLocale Current locale number Locale that was negotiated while collection occurred.
ISRV_Path Path string Complete PATH string for the operating system, which uses it
to search for application names when executed.
ISRV_Type 2048, 32768 The kind of system being managed. The number for UNIX,
Linux, and Xenix is 2048.
ISRV_DateTime Date time string Date that the file was generated.
DCH_CPUTypeList CPU types string Types of CPUs that this motherboard supports.
CG_Name Group name string Name of the domain or group that the system belongs to.
CGT_Name Group type name string Type of group. Has to be one of the valid group types.
ISR_SlotNumber RAM slot number Slot number that holds the RAM module.
ISC_Identity CPU identity string Unique number that identifies the CPU installed on the
system.
ISC_SlotNumber CPU slot number Slot number that holds the CPU.
ISC_SerialNumber Serial number string A serialized string of functionality. In Windows, this number
describes the feature set of the CPU.
DCPU_Family CPU family number Family number of the CPU. The Intel Pentium through
Pentium III are 6. The Pentium IV is 15.
DCPU_ModelNum CPU model number Model number of the CPU.
DCPU_Stepping CPU stepping number Represents any small change in CPU manufacturing.
DCPU_Model CPU model string The combined string of all the model numbers that describe
the processor.
DCPU_Rated Speed CPU rated speed number Designed maximum processor speed.
DCPU_Flags CPU flags string On Linux systems, the serial number in text form.
ISDA_Identity Drive adapter identity Unique numbe that identifies the drive adapter installed on the
string system.
ISDA_BusNumber Bus number Bus number where the drive adapter is positioned.
DDA_Type SCSI, IDE, SATA Drive adapter type string. Any string can be used, but for
consistency, you should try to use one of the specified values.
DDA_MaxSpeed Maximum speed number Maximum speed in MB/sec of the drive adapter interface.
ISDR_Identity Drive identity string Unique number that identifies the installed drive.
DDR_Space Drive size string Size of drive in GB as seen by the operating system.
DDR_Cache Drive cache size number Size of drive buffer cache in KBs.
DDR_TotalCyl Drive total cylinder Total number of cylinders that the drive reports to the
number operating system.
DDR_TotalHeads Drive total heads number Total number of heads that the drive reports to the operating
system.
DDR_TotalSectors Drive total sectors number Total number of sectors that the drive reports to the operating
system.
ISDR_CurIntSpeed Drive current interface Interface speed in MHz that was negotiated between the drive
speed number adapter and the drive. Normally the best speed of the slowest
component but can be overridden by the operating system to a
slower speed.
ISS_Identity Identity string Unique string that identifies the service or daemon on the
system.
ISS_IsActive Active bit (0 or 1) Indicator that the service or daemon is still running. Not
always possible on UNIX.
DSVC_DisplayName Name string Display name of service or daemon that is displayed in the UI.
DSVC_Name Name string Actual name of the service or daemon that the system
references. Usually unique without spaces.
ISN_Identity Identity string Identity string to uniquely identify the network adapter.
ISN_IPAddrType Static, dynamic String indicating whether the IP address is static or dynamic.
ISN_MACAddress MAC address string Media access control address for the network adapter.
ISN_CurSpeed NIC current speed Bandwidth in bits per second.
number
ISSH_NetName NetName string Share name that is published on the network. On UNIX, this is
the same name as the shared directory.
PERS_Type Type number Indicates the type of object that generated this performance
summary record.
Perfmon=0
Registry=1
WMI=2
Script=3
SNMP=4
PERS_Active Active flag bit Indicates that the summary counter is active. Inactive
performance objects are either grayed out or hidden in the
Data Manager.
PERS_ClassName Class name string Resembles a group name of similar performance data.
PERS_InstanceName Instance name string Represents distinct performance objects of the same class. If
the class does not have instances, the instance name can be left
blank or removed from the output.
PERS_MetricName Metric name string The actual system resource that the performance data
represents.
PERD_CounterAvg Sample average number Average of all the samples taken during collection.
The collector runs the scripts on the collector host, polling data from the discovered target machines
1 Log in to the target Linux or UNIX machine as root and create a directory under /tmp (or wherever you
want to place them) to hold the files. For example:
mkdir /tmp/vmware
NOTE The example shows files in the /tmp directory. If you change the location, use the changed
directory during the process.
2 Using a copy utility such as FTP, RCP, or SCP, copy the script files from C:\Program
Files\VMware\VMware Capacity Planner\scripts on the collector host (or an alternative installation
directory if you changed the default selection) to the directory on the target machine that you created
previously .
3 Change the file mode of aog_setup.sh to allow run privileges. For example:
chmod 755 /tmp/vmware/aog_setup.sh
4 Running the aog_setup.sh installs the scripts and creates the necessary cron entries. The files are:
aog_setup.sh—Creates directories and inserts the cron commands that start the following scripts in the
proper sequence
Either add a parameter to the end of the script or run the script without a parameter to use the directory
that you copied the scripts to in Step 1. The following are two examples:
/tmp/vmware/aog_setup.sh
/tmp/vmware/aog_setup.sh /opt/vmware
If the scripts prompt you for a missing script file titled aogconfigperf.sh, type the following:
touch /tmp/vmware/aogconfigperf.sh
5 If you installed the product sucessfully, the output appears similar to the following:
[/tmp/vmware]# crontab -l
30 23 1,15 * * /tmp/vmware/bin/aog_inv_cron.sh 0 * * * *
/tmp/vmware/bin/aog_perf_cron.sh
6 The scripts that are run according to the cron entries create inventory and performance files called output
in a subdirectory on each target machine called “output.”
Using ASCII format and CRLF, copy these files from the output directory on the target machine to the
import directory of the Capacity Planner collector host for processing. The default location of the import
directory is:
C:\Program Files\VMware\VMware Capacity Planner\import
During the next scheduled data synchronization job, the data synchronization process looks for files in the
Import directory. When files are found, the Data Manager creates CSV output files to be sent to the the
Information Warehouse.
The contents of the Import directory are deleted when the data synchronization occurs.
If errors are found in the imported data files, only the section of the file containing an error is rejected. Errors
are logged in the collector log file. By default, log files are stored in the following directory on the collector
host: C:\Program Files\VMware\VMware Capacity Planner.
CAUTION For each target machine, after the initial imports are completed, review the log files for any possible
errors.
1 Expand the navigation tree in the collector for each target machine to view CPU details.
2 View the information for the same target machine on the Dashboard to make sure that a complete
inventory is being sent to the Information Warehouse.
2 Remove the collection jobs from the cron file by using the following commands:
crontab -l | grep -v '/aog_' > /tmp/vcpcronremove
crontab /tmp/vcpcronremove
rm -f /tmp/vcpcronremove
3 Remove the script files and the directories created by them from the installation directory by using the
following command:
rm -frd /tmp/vmware
alert
An email message that indicates either a performance that exceeds vendor-provided thresholds, a
potential security risk, a change in the operating conditions of a monitored device, or an error condition.
anomaly
Indicates performance that is significantly different from the industry performance averages of like
servers that the Information Warehouse provides.
An anomaly is determined when CapacityPlanner detects performance that is more than three standard
deviations from the industry average. A performance difference an anomaly indicates can be either good
or bad. A good anomaly can provide a comparison to real-world, real-time performance that can be used
to configure similar machines, whereas a bad (or, undesirable) anomaly can identify a problem and allow
it to be addressed proactively before serious performance issues occur.
available bytes
Amount of physical memory available to processes running on a system. This parameter is calculated by
summing space on the zeroed, free, and stand-by memory list. Free memory is ready for use. Zeroed memory
is memory that is filled with zeros to prevent later processes from seeing data that a previous process
used. Standby memory is memory removed from a process working set (its physical memory) on the way
to disk, but that is still available to be recalled. This counter displays the last-observed value only; it is not
an average.
business hours
Represents, by default, the time period from 9:00 a.m. to 6:00 p.m. Also considered prime time.
redeployable servers. In addition, the Capacity Planner Dashboard provides “What if” analysis and
phantom server scenario analysis.
Consolidation Report
Provides a list of the successful consolidations that were generated. The report details the source system,
target system, and what their combined load looks like.
collector
Performs discovery, inventory, and performance functions, as well as the Collection, Export, and Send
functions of the Performance module. The collector runs as a Windows service. It requires local Windows,
Linux, or UNIX administrative rights to collect data specified systems, and collects data by using
Microsoft WMI (Windows Management Instrumentation) and Perfmon.
collector host
Client-side Windows-based system on which the collector and Capacity Planner Data Manager are
installed for client network analysis.
CPU megahertz
Real CPU speed, based on available processing capacity, not rated speed.
CPU use
Normalized number determined by the megahertz rate times the actual usage rating.
daemon
Program that is not invoked explicitly, but lies dormant waiting for some conditions to occur. The
perpetrator of the condition need not be aware that a daemon is lurking.
discovery
Process of searching a network for systems (servers or workstations) to be inventoried, which is
performed by the collector. The discovery process also discovers domains in the network. See also
inventory.
dynamic reports
Allow users to create custom reports. Users can select which type of report to view, and which fields to
include in that report. For any selected report, a user can search for particular values, add additional
criteria for sorting, and generate resulting reports that can be exported to a CSV file for analysis in Excel
or other database.
expert recommendations
By default, CapacityPlanner recommends the consolidation of two loads only if they are in the same
Department, Location, Environment, and Function. In addition, the combination of loads must not exceed
any of the specified thresholds.
CapacityPlanner collects many performance metrics and uses ten of these metrics to determine a
statistical fit. If the combination of loads exceeds any of these metrics, no consolidation recommendation
is made.
CapacityPlanner uses the prime time values when considering consolidation, not the weekly average, and
each server has its own load value. You can set different thresholds for different functional groups of
servers as well as override any of the set thresholds when you process recommendations.
firewall
Server with special security precautions on it, used to service outside network (especially Internet)
connections and dial-in lines.
group by
Filter that allows users to view performance based on Department, Environment, Function, and Location
groups as well as individual servers.
Information Warehouse
Contains industry performance averages for different types of servers, with information regarding
maximum observed values or thresholds on server resources. This intelligence provides benchmarking
information for anomaly detection, server consolidation road maps, and other analyses.
inventory
Process by which the collector collects information from every system (server or workstation) that was
identified during the discovery process. The information that the Inventory process can identify differs
from system to system, depending on the type of operating system platform, and the profile of the system
being inventoried. In addition, the Inventory process can be tailored to only retrieve specific types of
information from each system. See also Performance module.
IPC$
A hidden share used primarily to remotely administer network servers. See also network share.
multiprocessor balance
Illustrated by the CPU Balance Report; shows CPU usage on multiprocessor systems; identifies servers
where CPU processing is not evenly distributed across all processors.
network share
A location on a computer network that allows multiple computer users on the same network to share a
central space for storing files. See also IPC$.
obsolete server
Server that does not meet minimum CPU speed requirements, as defined by the Capacity Planner
Administrator.
page file
Component of an operating system that provides virtual memory for the system. Also called a swap file,
it is an area on disk where least recently used pages of memory are moved to, making room in physical
memory (RAM) for newer memory pages. The operation of exchanging memory pages between real
memory and the page file is called paging, or swapping, and is transparent to applications running on the
system platform.
However, the size and organization of the page file are major factors in the performance of the paging
operations and, thus, on entire system performance, so it is desirable to monitor the performance of the
paging operations with the page file on each server.
percentage CPU
Displays average percentage of busy time observed during the sample interval; calculated by monitoring
the time a server is inactive, then subtracting the value from 100%. This counter is a primary indicator of
processor activity.
Perfmon
Tool that allows performance statistics to be collected and accessed by user-level applications on
Windows systems.
Performance module
Monitors inventoried systems and collects information from those systems based on selected performance
counters. The collection process can be scheduled to occur at regular intervals over a specific period.
After performance data is collected, the Performance module exports the data and sends it to the Capacity
Planner Data Analyzer for incorporation in the Information Warehouse.
performance counters
A performance counter is a small amount of storage on a system that serves the special purpose of storing
the count of activity within that system. Performance data is collected by the Performance module.
performance week
Filter that allows users to view weekly average statistics collected over the last three years.
phantom server
A theoretical server that is placed in consolidation scenarios to see the affect on performance. Used as an
aid to making expert recommendations.
prime time
See cache hit percentage.
redeployable report
Provides a list of systems that are available for redeployment. These are systems that are not considered
obsolete, but have had their processing load moved to another system.
Registry
In Microsoft Windows, the Registry is a database repository for information about a computer's
configuration. The Registry contains information that Windows continually references during operation,
such as profiles for each user, programs installed on the computer and the types of documents each can
create, property settings for folders and program icons, what hardware exists on the system, and which
ports are being used.
The Registry is organized hierarchically as a tree and is made up of keys and their subkeys, and value
entries.
server group
Collection of servers based on their location, department, environment, function, domain or other group
characteristic.
stat sample
Filter that allows users to view information based on prime time average versus business hours.
swap file
See page file.
total pages
Number of pages in the buffer pool that includes database, free, and stolen.
trend deviation
Indicates when a server has deviated from the trending line on a particular performance statistic (which
can result in a shortened server life or major performance issue).
trends
Trending report that can be customized by the counter that you want to view. All trending information
can also be filtered by groups.
trusted domain
Directory domain that shares user rights and privileges with another domain.
UDP
Protocol in the TCP/IP protocol suite, the User Datagram Protocol or UDP allows an application program
to send datagrams to other application programs on a remote machine. UDP is a protocol that provides
an unreliable and connectionless datagram service where delivery and duplicate detection are not
guaranteed. It does not use acknowledgments, or control the order of arrival.
virtual memory
Extension of a system’s physical memory, enabled by the declaration of a page file.
overview 17 refreshing
Perfmon 95 inventory data 93
timing 99 regedit.exe 124
performance collection timing 99 regedt32.exe 124
performance counters registration
defined 139 Dashboard 16
derived 160 Registry connection 17
removing 99 Registry Editor 124
performance data missing 122 Registry inventory collection 89
Performance reports menu 63 Registry Only option problems 121
performance tuning 18 Remote Registry and Perfmon 88
personnel for an assessment 22 Remote Registry service 121
phases of an assessment 21 removing
Physical Disk Derived performance counters 160 data files 134
Physical Disk performance counters 153 performance counters 99
Physical Disk Utilization performance report 100 removing Capacity Planner 134
ping command 123 Reports 105
pinging objects 108 Reports menu 62
platforms supported 28 reports overview 114
pre-assessment tasks 22 requirements
pre-defined scheduled jobs 109 Collector host system 27
pre-installation checklist 31 Collector service
pre-installation tasks 22 Dashboard 16
Print Queue performance counters 154 host system account 29
privileges inventory collection 88
Collector accounts 29 Linux and UNIX 29
problems Microsoft Active Directory 68
connecting to the IP address 118 network connection 30
name resolution 117 network connections 18
with installation 40 security and network 29
procedure system access account 44
discover domains 70 UNIX and Linux 48
process user accounts 27
assessment 21 reset collection 93
discovery 16 resolving the host name 119
inventory data collection 17 resolving unknown objects 79
Process performance counters 154 resources
Processor performance counters 155 troubleshooting 115
Processor Queue Length performance report 99 Run Manual Tasks menu 62
Processor Utilization performance report 99 Run Scheduled Jobs menu 62
Purge Database command 126 Run Setup Wizard 65
purpose for assessment 21 running
PuTTY application 80, 124 Collector service 40
documentation 49, 80 scheduled jobs manually 110
shareware 48 UNIX and Linux scripts 113
R S
reasons for assessment 21 saving
RedHat Linux systems supported 28 collection time 108
Redirector performance counters 157 log files 107
rediscover performance option 97 reports 114
reducing collection time 108 schedule
Collector jobs 109