system monitor pms user's guide1 chapter 1 functions 1.1 function overview of system monitor -...
TRANSCRIPT
User's Guide
System Monitor - Performance Monitoring Services 5.4
Preface
System Monitor - Performance Monitoring Services 5.4 (hereafter referred to as "System Monitor - Performance
Monitoring Services") is a software program that provides a simple means of monitoring the performance of
multiple machines. This manual explains how to use System Monitor - Performance Monitoring Services.
This manual is organized as follows. Chapter Title Content
1 Function Overview The functions and features of System Monitor - Performance Monitoring Services
2 Configuration setting How to make configuration settings 3 Basic Operations Startup methods and basic GUI operations 4 Collecting Performance Data How to make settings relating to collecting performance data
5 Displaying Performance Status
How to make settings relating to displaying performance status
6 Threshold Monitoring and the Notification Function
How to monitor thresholds for performance data and make settings for the notification function
7 Management of Performance Data
How to manage performance data
8 Command-Line Interface How to use command-line interface 9 Troubleshooting Errors and how to handle them 10 Notes Various notes and issues
First edition, January 2004
20th edition, April 2014
Notes
(1) No part of this document may be reproduced in any form without the prior written permission of NEC
Corporation.
(2) The contents of this document may be revised without prior notice.
(3) All efforts have been made to ensure the accuracy of all information in this document. If you notice any part
unclear, incorrect, or omitted in the document, contact your authorized NEC sales representative.
(4) NEC assumes no liability for damages arising from the use of this product, nor any liability for incidental or
consequential damages arising from the use of this document regardless of (3)
Trademarks • "Windows", "Hyper-V", "Microsoft SQL Server" and "Microsoft SQL Server 2012 Express Edition" are
registered trademarks or trademarks of Microsoft Corporation, U.S. in the United States and other
countries.
• "Linux" is a registered trademark or trademark of Linus Torvalds in the United States and other countries.
• "Red Hat" is a registered trademark or trademark of Red Hat, Inc. in the United States and other countries.
• "SUSE" is a trademark of Novel, Inc.
• "VMware" is a registered trademark or trademark of VMware, Inc.
• Citrix XenServer is a registered trademark or trademark of Citrix Systems, Inc.
• Other product names that appear in this document may be registered trademarks or trademarks of their
respective owners.
This product includes icons from the Silk Icons set version 1.3 by Mark James licensed under a Creative
Commons Attribution 2.5 License.
http://www.famfamfam.com/
This product includes Granados provided by Routrek Networks, Inc under the following license:
Copyright (c) 2003 Routrek Networks, Inc. All Rights Reserved.
Copyright (c) 2002 Chew Keong TAN
Redistribution and use in source and binary forms, with or without modification, are permitted provided
that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the
following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the
following disclaimer in the documentation and/or other materials provided with the distribution.
3. The end-user documentation included with the redistribution, if any, must include the following
acknowledgment:
"This product includes software developed by Routrek Networks, Inc."
Alternately, this acknowledgment may appear in the software itself,
if and wherever such third-party acknowledgments normally appear.
4. The names "Routrek Networks" or "Routrek" must not be used to endorse or promote products derived
from this software without prior written permission. For written permission, please contact
5. Products derived from this software may not be called "Granados", appear in their name, without prior
written permission of the Routrek Networks, Inc.
THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL ROUTREK
NETWORKS, INC. OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE
OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Copyright (c)NEC Corporation 2004,2014
This document cannot be changed or copied without permission from NEC.
Table of Contents
CHAPTER 1 FUNCTIONS ..................................................................................................................... 1
1.1 FUNCTION OVERVIEW OF SYSTEM MONITOR - PERFORMANCE MONITORING SERVICES ................... 1 1.2 BASIC CONFIGURATION OF SYSTEM MONITOR - PERFORMANCE MONITORING SERVICES ................. 2 1.3 COLLECTED PERFORMANCE DATA ..................................................................................................... 7 1.4 SUMMARY DATA ............................................................................................................................... 19 1.5 DATA MANAGEMENT ........................................................................................................................ 20 1.6 GROUPS ............................................................................................................................................ 21 1.7 CONNECTING TO MONITORED MACHINES ....................................................................................... 22 1.8 DISPLAYING PERFORMANCE STATUS AND CSV FILE OUTPUT ......................................................... 29 1.9 THRESHOLD MONITORING AND THE NOTIFICATION FUNCTION ....................................................... 36 1.10 THE SYSTEMPROVISIONING LINKAGE FUNCTION ........................................................................ 39 1.11 COMMAND-LINE INTERFACE......................................................................................................... 64
CHAPTER 2 CONFIGURATION SETTINGS ....................................................................................... 66
2.1 STARTING THE PERFORMANCE MONITORING SERVICE .................................................................... 66 2.2 REGISTERING THE MANAGEMENT SERVER ...................................................................................... 67 2.3 CONNECTION SETTINGS TO MONITORED MACHINES ....................................................................... 68 2.4 CONNECTION SETTINGS TO THE MANAGEMENT CONSOLE .............................................................. 69 2.5 THE EXECUTION ACCOUNT FOR THE PERFORMANCE MONITORING SERVICE .................................. 72 2.6 SYSTEMPROVISIONING CONNECTION SETTINGS.............................................................................. 76 2.7 LOG SETTINGS ................................................................................................................................. 78
CHAPTER 3 BASIC OPERATIONS ..................................................................................................... 79
3.1 STARTING THE MANAGEMENT CONSOLE AND CONNECTING TO THE MANAGEMENT SERVER .......... 79 3.2 THE MAIN WINDOW ......................................................................................................................... 83
CHAPTER 4 COLLECTING PERFORMANCE DATA ......................................................................... 89
4.1 SPECIFICATIONS FOR MONITORED MACHINES ................................................................................. 89 4.2 SPECIFICATIONS FOR THE PERFORMANCE DATA TO BE COLLECTED .............................................. 100 4.3 PAUSING AND RESTARTING THE COLLECTING OF PERFORMANCE DATA ........................................ 113 4.4 CHECKING THE COLLECTING STATUS FOR PERFORMANCE DATA .................................................. 114
CHAPTER 5 PERFORMANCE STATUS DISPLAY AND FILE OUTPUT ......................................... 117
5.1 DISPLAYING PERFORMANCE GRAPHS ............................................................................................ 117 5.2 SAVING PERFORMANCE STATUS DISPLAY SPECIFICATIONS ........................................................... 130 5.3 FILE OUTPUT ................................................................................................................................. 131
CHAPTER 6 THRESHOLD MONITORING AND NOTIFICATION .................................................. 133
6.1 CONFIGURING THRESHOLD MONITORING ..................................................................................... 133
CHAPTER 7 MANAGEMENT OF PERFORMANCE DATA .............................................................. 143
7.1 DATA MANAGEMENT TOOL ............................................................................................................ 143 7.2 DATABASE SIZE MONITORING ........................................................................................................ 146
CHAPTER 8 COMMAND-LINE INTERFACE ................................................................................... 148
8.1 SSC-PERF COMMAND ...................................................................................................................... 148
CHAPTER 9 TROUBLESHOOTING .................................................................................................. 154
9.1 THE EVENT LOG ............................................................................................................................ 154 9.2 WHAT TO DO IF THE PERFORMANCE MONITORING SERVICE DOES NOT START .............................. 161 9.3 WHAT TO DO IF THE CONNECTION FROM MANAGEMENT CONSOLE TO SERVER FAILS ................... 167 9.4 WHAT TO DO IF PERFORMANCE DATA COLLECTING FAILS ............................................................. 169 9.5 WHAT TO DO IF PERFORMANCE DATA COLLECTING IS DELAYED ................................................... 175 9.6 WHAT TO DO IF THERE ARE SYSTEMPROVISIONING CONFIGURATION APPLYING ERRORS ............. 176 9.7 WHAT TO DO IF THERE ARE ERRORS WHEN PERFORMANCE ABNORMALITIES ARE REPORTED TO
SYSTEMPROVISIONING ............................................................................................................................ 177 9.8 WHAT TO DO IF THE SYSTEMMONITOR DATA MANAGEMENT TOOL MIGHT NOT START .................... 178 9.9 DATABASE ...................................................................................................................................... 180 9.10 OTHER ........................................................................................................................................ 184
CHAPTER 10 NOTES ......................................................................................................................... 185
10.1 THE NUMBER OF MONITORED MACHINES .................................................................................. 185 10.2 NOTES ON DATABASE .................................................................................................................. 186 10.3 NOTES ON GRAPH DISPLAYS ....................................................................................................... 187 10.4 NOTES ON LINKING TO SYSTEMPROVISIONING .......................................................................... 188 10.5 RELEASE OF RESOURCES ............................................................................................................ 193 10.6 NOTES ON UPGRADING/UPDATING ............................................................................................. 194 10.7 ISSUES WHEN OTHER PRODUCTS TO MONITOR THE WINDOWS MONITORED MACHINE COEXIST
WITH SYSTEM MONITOR - PERFORMANCE MONITORING SERVICES ......................................................... 195
APPENDIX .......................................................................................................................................... 196
A. HOW PERFORMANCE VALUES FOR GROUPS ARE CALCULATED ........................................................ 196
1
Chapter 1 Functions
1.1 Function Overview of System Monitor - Performance Monitoring Services
System Monitor - Performance Monitoring Services is a software program that supports monitoring the
performance status of systems, and it provides features to view the performance state and to monitor thresholds
by periodically collecting performance data from the system and accumulating those data.
You can use the methods of collecting performance data provided by a monitored machine's operating system or
virtualized infrastructure from the physical machine running on Windows or Linux, or from the virtual
machine server or virtual machine running on VMware, Hyper-V, Xen Server or KVM. Alternatively, you can
execute an arbitrary script on a management server or monitored machine to collect performance data.
You can efficiently keep the collected performance data for a long period because System Monitor - Performance
Monitoring Services aggregates the data as “summary data” and accumulates those aggregated data.
System Monitor - Performance Monitoring Services provides graph view for the collected performance data
which makes it easy to keep track of the performance status of the system.
For systems with large-scale configurations, the performance status of the entire system can be followed at a
glance by using the function that groups machines and graphs the performance status of each group. You can
view the graph of the performance status on the SigmaSystemCenter Web console or the SystemMonitor
management console.
Performance problems (such as high load) can be discovered quickly by graphically displaying and monitoring
the performance status of the system in real time.
It is also possible to display performance data that has been collected in the past again. This enables the
operational status of the system in the past to be checked.
The threshold monitoring feature can automatically compare the collected performance data and an arbitrary
threshold, and, in case of exceeding threshold, it can log unusual loads on the monitored machine to the event
log or notify SystemProvisioning of it. In notifying SystemProvisioning, SigmaSystemCenter can run the policy
action in response to the notified events.
To make the VM Optimized Placement feature available, which is the feature for load balancing or power saving
in the virtual environment of SigmaSystemCenter, you must use the threshold monitoring feature of System
Monitor - Performance Monitoring Services.
You can use the monitoring profile settings from the SigmaSystemCenter Web console to configure collecting
performance data or thresholds on System Monitor - Performance Monitoring Services. However, if the
configuration requires any special customization, use the System Monitor - Performance Monitoring Services
management console.
2
1.2 Basic Configuration of System Monitor - Performance Monitoring Services
System Monitor - Performance Monitoring Services is one of the SigmaSystemCenter products, and it consists
of a "Performance Monitoring Service function" that collects performance data and a "management console
function" that is the user interface for using the Performance Monitoring Service. Typically, System Monitor -
Performance Monitoring Services is used in combination with SystemProvisioning.
1 minute interval 5 minutes 15 minutes
1 day 1 hour
Aggrega-ting
System Monitor - Performance Monitoring Services
Management console
System Monitor - Performance Monitoring Services
Performance Monitoring Service Applying configuration information (such as a monitoring profile)
Making settings
Displaying graphs
Saved for 3 days
1 week 1 month 3 months 5 years
Accumulates
performance
data
Management server
Obtaining the data for Web graph display
Notifying exceeding threshold
SystemProvisioning
Web graph display
Setting a monitoring profile
Running policy actions Obtaining
performance data
Monitored machines
Aggrega- ting
Aggrega- ting
Aggrega-ting
3
(1) The Performance Monitoring Service
This function resides on the server, and acquires performance data by accessing the monitored machines.
The performance data is collected and stored in a database. Keeping the collected performance data for a
long period requires a large data space in a database. However, System Monitor - Performance Monitoring
Service can aggregate the data to one collection of the data in a longer interval and accumulate the data
preventing the data area from filling rapidly. The machine where the Performance Monitoring Service runs
is called the "management server" and the machine whose performance is monitored should be called the
"monitored machine".
The Performance Monitoring Service runs in the background as a Windows service and starts
automatically when the OS starts.
(2) The Management Console
This is one of the user interfaces for using the Performance Monitoring Service. Settings for the
Performance Monitoring Service, such as which machines are monitored and which performance
information is collected, are performed from the management console. The management console can also
graphically display performance data. If you intend to use not System Monitor - Performance Monitoring
Services but the SigmaSystemCenter features to manage a monitored machine, use the
SigmaSystemCenter Web console (described later) instead of the System Monitor - Performance
Monitoring Services management console.
The management console is installed on the management server along with the Performance Monitoring
Service. Users connect to the Performance Monitoring Service by starting the management console on the
management server. More than one management console can connect to the same Performance Monitoring
Service at the same time.
In addition, central management is available for multiple management servers from one management
console. To implement performance monitoring on large-scale systems, separate monitored machines into
the appropriate units and prepare a management server for each.
4
(3) SystemProvisioning/SigmaSystemCenter Web console
SystemProvisioning is a core product of SigmaSystemCenter and performs configuration management and
provisioning for management machines in SigmaSystemCenter. To perform operations for configuring or
provisioning management machines, use the SigmaSystemCenter Web console.
You can use the monitoring profile information (the set of performance indicator settings and thresholds
monitoring settings) on the SigmaSystemCenter Web console to configure a monitored machine of System
Monitor - Performance Monitoring Services and a performance indicator to be collected.
System Monitor - Performance Monitoring Services applies the monitoring profile information and the
configuration information of SystemProvisioning to itself at the timing of execution of reflecting the
configuration information of SystemProvisioning. This allows System Monitor - Performance Monitoring
Services to automatically start performance monitoring for the monitored machine that you intend to
register and allows to display the collected performance data in graph on the SigmaSystemCenter Web
console.
System Monitor - Performance Monitoring Services can notify SystemProvisioning of performance
problems detected by it such as exceeding thresholds. This makes it possible to automatically start the
policy action configured on SystemProvisioning.
One SystemProvisioning can work with more than one System Monitor – Performance Monitoring Services.
In performance monitoring in the large-scale system environment, monitored machines can be grouped by
Monitoring Target Range of Management Console
Management server1
System Monitor - Performance Monitoring
Services
Monitored machines
System Monitor - Performance Monitoring Service
Management console
Management server2
System Monitor - Performance Monitoring
Services
Management server3
System Monitor - Performance Monitoring
Services
5
group or model in SystemProvisioning and each monitored machine can be monitored on a management
server. In this case, you can view the performance information for the monitored machine in graph on the
Web console of each management server.
Operating Systems
Supported operating systems for the management server and monitored machines are as follows:
Supported operating systems
Management server Windows Server 2008 R2 Standard Edition SP1 Windows Server 2008 R2 Enterprise Edition SP1 Windows Server 2008 R2 Datacenter Edition SP1 Windows Server 2012 Standard Edition Windows Server 2012 Datacenter Edition Windows Server 2012 R2 Standard Edition Windows Server 2012 R2 Datacenter Edition
Monitored machines Windows 2000 Server (SP4 or later) Windows 2000 Advanced Server (SP4 or later) Windows Server 2003, Standard Edition (including SP1, R2, SP2, and R2 SP2) Windows Server 2003, Enterprise Edition (including SP1, R2, SP2, and R2 SP2) Windows Server 2003, Standard x64 Edition (including R2, SP2, and R2 SP2) Windows Server 2003, Enterprise x64 Edition (including R2, SP2, and R2 SP2) Windows Server 2008 Standard Edition (including SP2) Windows Server 2008 Enterprise Edition (including SP2) Windows Server 2008 Standard x64 Edition (including SP2) Windows Server 2008 Enterprise x64 Edition (including SP2) Windows Server 2008 Datacenter x64 Edition (including SP2) Windows Server 2008 R2 Standard Edition (including SP1) Windows Server 2008 R2 Enterprise Edition (including SP1) Windows Server 2008 R2 Datacenter Edition (including SP1) Windows Server 2012 Standard Edition Windows Server 2012 Datacenter Edition Windows Server 2012 R2 Standard Edition Windows Server 2012 R2 Datacenter Edition Windows XP Professional (including SP1, SP2, SP3) Windows Vista Business Edition (including SP1, SP2) Windows Vista Enterprise Edition (including SP1, SP2) Windows Vista Ultimate Edition (including SP1, SP2) Windows 7 Professional Edition (including SP1) Windows 7 Enterprise Edition (including SP1) Windows 7 Ultimate Edition (including SP1) Windows 8 Pro Edition Windows 8 Enterprise Edition Windows 8.1 Pro Edition Windows 8.1 Enterprise Edition Red Hat Enterprise Linux ES/AS 4 (including x64)
6
When Windows Vista or Windows Server 2008 is running on the monitored
machine, it may take longer time to collect performance data because of the
internal behavior of Windows. For detail, see "9.5 What to do if Performance
Data Collecting is Delayed."
The version 4.8 and earlier is evaluated for Red Hat Enterprise Linux ES/AS
4 as a monitored machine. The version 5.4 and earlier is evaluated for Red
Hat Enterprise Linux 5 Server as a monitored machine. The version 6.5 and
earlier is evaluated for Red Hat Enterprise Linux 6 Server as a monitored
machine. For the later version, please confirm our latest evaluation status.
Service Pack 2 is necessary for collecting performance data from Windows
Server 2008 Standard/Enterprise x64 Hyper-V.
Guest operating systems supported by VMware ESX, Citrix XenServer,
Hyper-V, and KVM are supported only if they are supported as operating
systems for monitored machines in System Monitor - Performance
Monitoring Services.
Supported operating systems Red Hat Enterprise Linux 5, 6 Server (including x64) SUSE Linux Enterprise Server 9 (SP3), 10 (SP1, SP2) VMware ESX 4.0, 4.1 VMware ESXi 4.0, 4.1, 5.0, 5.1, 5.5 Citrix XenServer 5.6, 6.0 Windows Server 2008 R2 Standard Hyper-V (including SP1) Windows Server 2008 R2 Enterprise Hyper-V (including SP1) Windows Server 2012 Hyper-V Windows Server 2012 R2 Hyper-V Red Hat Enterprise Linux 6.5 KVM Guest operating systems on VMware ESX/ESXi Guest operating systems on Citrix XenServer Guest operating systems on Hyper-V Guest operating systems on KVM
Note
Note
Note
Note
7
1.3 Collected Performance Data
System Monitor - Performance Monitoring Services periodically collects performance data from the monitored
machines. Collected Performance Indicators can be specified for each SystemMonitor management server, for
each group or for each machine. If they are specified for a management server, they are shared with all
monitored machines that are monitoring targets of the management server.
You can configure to collect performance data for the monitored machine on the monitoring profile of
SigmaSystemCenter or on the System Monitor – Performance Monitoring Services management console.
Configuring on the monitoring profile of SigmaSystemCenter allows you to configure the monitored machine
registered by reflecting the configuration information of SystemProvisioning. The specified performance
indicator on the monitoring profile of SigmaSystemCenter is automatically applied to System Monitor -
Performance Monitoring Services at the timing of reflecting the configuration information of
SystemProvisioning. For how to configure the monitoring profile, see 4.11.3, "How to Create a Monitoring
Profile" and 5.5.15, "Configuring Settings on the Performance Monitor Tab" in SigmaSystemCenter
Configuration Guide.
For how to configure it on the management console of System Monitor – Performance Monitoring Services, see
4.2.2, "How to Specify Collecting Data Settings"
1.3.1 Types of Performance Indicators and Collecting Intervals
The Performance Indicators consist of "Resource", "Performance index" and "Instance". "Resource" is the target
resource of the Performance Indicators, and includes CPU, physical disk and memory. "Performance index" is
the type of data collected, and indicates CPU usage and disk transfer rates. "Instance" identifies a particular
measurement object in situations where a resource contains multiple measurement objects. "Instance" is unique
to "Performance index". The settings for "Instance" cannot be changed. Up to 256 items of Performance
Indicators can be specified.
Performance data is collected from monitored machines at regular intervals. These intervals are called
"collecting intervals", and can be specified individually for each Performance Indicator item or collectively for all
Performance Indicator items.
Performance Indicators that have been configured to collect data, begin collecting from monitored machines
automatically when the Performance Monitoring Service starts.
1.3.2 Built-in Performance Indicator
System Monitor - Performance Monitoring Services comes with Built-in Performance Indicators that are
common to all systems. Some typical Built-in Performance Indicator items are set up as default settings for each
management server when System Monitor - Performance Monitoring Services is installed.
With built-in performance indicators, the individual data per device in the monitored machine cannot be
collected because built-in performance indicators collect performance data per monitored machine basis. If you
need to collect individual performance data per device basis, use custom performance indicators instead.
8
For example, the performance data obtained with Network Packet Transfer Rate (Bytes/sec) is the total rate of
all packets transferred through network adapters if a monitored machine has more than one network adapters.
(1) Standard Performance Indicator
Performance data for these performance indicators are collected by accessing the monitored machine
directly.
Resource Performance index ID Explanation
OS of monitored machine
Window
s
Linux
ESX/ESXi
Xen
Hyper-V
KVM
CPU CPU Usage (%) 1 CPU Usage (%) is the percentage of elapsed time that the processor spends to execute a non-Idle thread.
x x x x x x
CPU System Usage (%)
2 CPU System Usage (%) is the percentage of elapsed time that the process threads spent executing code in privileged mode.
x x - - - x
CPU User Usage (%)
3 CPU User Usage (%) is the percentage of elapsed time the processor spends in the user mode.
x x - - - x
CPU Usage (MHz) 4 CPU Usage (MHz) is the amount of used CPU resource that is calculated by CPU Usage (%) × the count of logical processors × CPU frequency.
x x x x x x
Disk Disk Transfer Rate (Bytes/sec)
21 Disk Transfer Rate (Bytes/sec) is the rate bytes are transferred to or from the disk during write or read operations.
x x x - x x
Disk IO Count (IO/sec)
22 Disk IO Count (IO/sec) is the rate of read and write operations on the disk.
x x x - x x
Disk Read Transfer Rate (Bytes/sec)
23 Disk Read Transfer Rate (Bytes/sec) is the rate at which bytes are transferred from the disk during read operations.
x x x - x x
Disk Read Count (IO/sec)
24 Disk Read Count (IO/sec) is the rate of read operations on the disk.
x x x - x x
Disk Write Transfer Rate (Bytes/sec)
25 Disk Write Transfer Rate (Bytes/sec) is the rate at which bytes are transferred to the disk during write operations.
x x x - x x
Disk Write Count (IO/sec)
26 Disk Write Count (IO/sec) is the rate of write operations on disk.
x x x - x x
Disk Space (MB) 27 Disk Space (MB) displays the unallocated space, in megabytes, on the disk drive. One megabyte is equal to 1,048,576 bytes.
x x x x x x
Disk Space Ratio (%)
28 Disk Space Ratio (%) is the percentage of total usable space on the whole disk drive that was free.
x x x x x x
9
If the OS for the monitored machine is Windows 2000, users must activate
the "LogicalDisk" counters manually to collect this performance data. Enter
"diskperf -y" on the monitored machine and restart the monitored machine
after executing the command. Also, stop and restart the Performance
Monitoring Service.
CPU Usage (MHz) is available only for the physical machine whose
configuration information is applied from SystemProvisioning.
Physical machines (except machines which run ESX/ESXi, Hyper-V or Citrix
Xen Server) must be automatically registered to NEC ESMPRO Manager by
SystemProvisioning.
The performance data for disk resource of Hyper-V does not include the
performance data of CSV (Cluster Shared Volume).
Network Network Packet Transfer Rate (Bytes/sec)
41 Network Packet Transfer Rate (Bytes/sec) is the rate at which data is transmitted and received across the all physical NICs. The value is total of Network Packet Reception Rate (Bytes/sec) and Network Packet Transmission Rate (Bytes/sec).
x x x x x x
Network Packet Reception Rate (Bytes/sec)
42 Network Packet Reception Rate (Bytes/sec) is the total rate at which data is received across the all physical NICs.
x x x x x x
Network Packet Transmission Rate (Bytes/sec)
43 Network Packet Transmission Rate (Bytes/sec) is the total rate at which data is transmitted across the all physical NICs.
x x x x x x
Memory Physical Memory Space (MB)
61 Physical Memory Space (MB) is the amount of physical memory, in Megabytes, immediately available for allocation to a process or for system use.
x x x x x x
Physical Memory Space Ratio (%)
62 Physical Memory Space Ratio (%) is the percentage of the available memory size to the installed physical memory size that is calculated by Physical Memory Space (MB) / the installed physical memory size × 100.
x x x x x x
Note
Note
Note
10
When the Dynamic Memory feature of the monitored virtual machines on
Hyper-V is enabled, the amount of Physical Memory Space Ratio (%) for
these virtual machines may be incorrect.
Physical Memory Space Ratio (%) for a monitored Windows/Hyper-V machine
is available only for the machine whose configuration information is applied
from SystemProvisioning.
Also, physical machines on Windows must be automatically registered to
NEC ESMPRO Manager by SystemProvisioning.
The performance data collected from Citrix XenServer consists of data
obtained in the last several seconds before the performance data is collected.
In the KVM monitored machine, each value of Disk Space (MB) and Disk
Space Ratio (%) is calculated as the total of the all storage pools on the KVM
host. If multiple storage pools share the same disk space, the Disk Space
(MB) and Disk Space Ratio (%) are incorrectly calculated because the shared
disk spaces are redundantly calculated. When using storage pools, avoid that
they share the same disk spaces.
For example, the following use cases may cause issues in calculation of Disk
Space and Disk Space Ratio:
1. Multiple directories on the same local file system are configured as
individual storage pools.
2. Multiple directories on the same file system in the same NFS server are
configured as individual storage pools.
(2) Performance Indicator for Virtual Machine
These are performance indicators only for a virtual machine.
If the monitored machine is a virtual machine running on a VMware ESX/ESXi, the ESX/ESXi virtual
machine server where the monitored virtual machine is running also must be monitored because
performance data of the virtual machine are collected via VMware ESX/ESXi. Also, the information of
VMware ESX/ESXi on which the monitored virtual machine is running is obtained from
SystemProvisioning, which requires a single SigmaSystemCenter management server to contain VMware
ESX/ESXi and the monitored virtual machine in the active state.
If the monitored machine is a virtual machine running on Citrix XenServer, Hyper-V or KVM, the
performance data of the virtual machine are obtained via SystemProvisioning; this requires the
information of the monitored virtual machine to be propagated to System Monitor-Performance Monitoring
Note
Note
Note
Note
11
Service by the SystemProvisioning configuration information reflecting feature. Because of this, the
monitored virtual machine in the active state must be registered to SigmaSystemCenter.
Resource Performance index ID Explanation
Virtualization Platform
ESX/ESXi
Xen
Hyper-V
KVM
CPU Guest CPU Usage (%)
11 Guest CPU Usage (%) is the percentage of the total CPU resource that actively used by the VM to the available CPU resource for the VM.
x x x x
Guest CPU Usage (MHz)
12 Guest CPU Usage (MHz) is the total CPU resource that actively used by the VM.
x x x x
Host CPU Usage (%)
13 Host CPU Usage (%) is the percentage of the VM server CPU resource that actively used by the VM to the total CPU resource of the VM server.
x - x x
Host CPU Usage (MHz)
14 Host CPU Usage (MHz) is the VM server CPU resource that actively used by the VM.
x - x x
Disk Guest Disk Transfer Rate (Bytes/sec)
31 Guest Disk Transfer Rate (Bytes/sec) is the rate at which data is transferred to or from the virtual disk on the VM during write or read operations.
x - x -
Guest Disk IO Count (IO/sec)
32 Guest Disk IO Count (IO/sec) is the I/O count per second when data is transferred to or from the virtual disk on the VM during write or read operations.
x - - -
Guest Disk Read Transfer Rate (Bytes/sec)
33 Guest Disk Read Transfer Rate (Bytes/sec) is the rate at which data is transferred from the virtual disk on the VM during read operations.
x - - -
Guest Disk Read Count (IO/sec)
34 Guest Disk Read Count (IO/sec) is the I/O count per second when data is transferred from the virtual disk on the VM during read operations.
x - - -
Guest Disk Write Transfer Rate (Bytes/sec)
35 Guest Disk Write Transfer Rate (Bytes/sec) is the rate at which data is transferred to the virtual disk on the VM during write operations.
x - - -
Guest Disk Write Count (IO/sec)
36 Guest Disk Write Count (IO/sec) is the I/O count per second when data is transferred to the virtual disk on the VM during write operations.
x - - -
Guest Disk Usage (MB)
37 Guest Disk Usage (MB) is the datastore size consumed by the VM. If the target VM is running on VMware ESX, this is the used space of the virtual disk allocated to the VM.
x x x x
Guest Disk Usage (%)
38 Guest Disk Usage (%) is the percentage of the datastore size consumed by the VM to the size of the allocated virtual disk. If the target VM is running on VMware ESX, this is the percentage of the used space of the virtual disk allocated to the VM.
x x x x
Network Guest Network Transfer Rate (Bytes/sec)
51 Guest Network Transfer Rate (Bytes/sec) is the rate at which data is transmitted and received across the all virtual NICs on the VM.
x - x -
Memory Guest Memory 71 Guest Memory Usage (%) is the percentage of x x x -
12
Collecting the performance information of a virtual machine is available only
for the virtual machine registered by the SystemProvisioning configuration
information reflecting feature.
When the monitored machine is a virtual machine running on a VMware
ESX/ESXi and performance data for Guest Disk Usage (%) or Guest Disk
Usage (MB) is collected, it is necessary that VMware Tools is running on the
monitored virtual machine.
To collect performance data for Guest Memory Usage (%) or Guest Memory
Usage (MB) from a virtual machine running on Hyper-V, Dynamic Memory
for the virtual machine must be enabled.
The performance data for a virtual machine collected from Citrix XenServer,
Hyper-V or KVM are the data obtained in the last several seconds before the
performance data is collected regardless of the collecting interval.
Usage (%) the memory size used by the VM to the memory size allocated to the VM. If the target VM is running on VMware ESX, this is the percentage of the size of the VM server memory used actively by the VM.
Guest Memory Usage (MB)
72 Guest Memory Usage (MB) is the memory size used by the VM. If the target VM is running on VMware ESX, this is the size of the VM server memory actively used by the VM.
x x x -
Host Memory Usage (%)
73 Host Memory Usage (%) is the percentage of the size of the VM server memory consumed for the VM to the total memory size of the VM server.
x x x -
Host Memory Usage (MB)
74 Host Memory Usage (MB) is the size of the VM server memory consumed for the VM.
x x x -
(3) Performance Indicator for Physical Machine
These are performance indicators only for a physical machine.
It is necessary for collecting performance data for this indicator that the configuration information of the
monitored physical machine is applied from SystemProvisioning. And it is also necessary that
configurations to make the OOB Management feature available for the monitored machine on
SigmaSystemCenter.
Note
Note
Note
Note
13
Resource Performance index ID Explanation
Power Supply Current Power (W) 101 Current Power (W) is the current power consumption of a physical machine.
1.3.3 Custom Performance Indicator
System Monitor-Performance Monitoring Service can collect the performance data of performance indicators
which are not default built-in performance indicators by defining them as custom performance indicators. The
following describes about the type of the custom performance indicator that you can define in System
Monitor-Performance Monitoring Services.
(1) Performance counters
For a Windows monitored machine
You can define a custom performance indicator by specifying the same categories, counters, and
instances for the Windows monitored machine as those in Windows Performance Monitor. The custom
performance indicator enables System Monitor - Performance Monitoring Services to collect the same
performance information of the Windows machine as that collected by Windows Performance Monitor.
For a VMware ESX/ESXi virtual machine server / virtual machine
When the monitored machine is the VMware ESX/ESXi virtual machine server or virtual machine, you
can define the custom performance indicator by specifying "metric group", "counter", and "object" like
you do in the Customize Performance Chart dialog box in VMware vSphere Client. The custom
performance indicator enables System Monitor - Performance Monitoring Services to collect the same
performance information of the VMware ESX/ESXi virtual machine server or virtual machine as that
collected by vSphere Client.
(2) User scripts
The user script has a capability to collect its output as a performance data by executing an arbitrary script.
The user script has two types: one is a remote user script which is executed on a monitored machine;
another is a local user script which is executed on a management server.
Remote user script
In a Windows / Linux monitored machine, you can create custom performance indicators by specifying
a script on the monitored machine. When collecting the performance data, System Monitor -
Performance Monitoring Services remotely connects to the monitored machine using SSH, and
executes the script on the monitored machine at every collecting interval. Then, System Monitor -
Performance Monitoring Services extracts the results from the [standard output] information to
obtain the performance data.
14
When creating the custom performance indicator, you must specify the name of the script file on the
monitored machine as the executing string. You can add the string "remote+ssh://" before the
executing string. Alternatively, you can specify the executing the string directly. Use the following
execution examples by script type as a reference. We recommend that you use the full path to specify
the script path to avoid execution errors.
For Windows
In Windows monitored machines, you can specify scripts such as batch files, Windows Script Host
(WSH) and PowerShell on the monitored machine.
• Batch file
Specify the batch file path.
E.g.) remote+ssh://C:\PerformanceMonitor\Script\test.bat
• WSH (Windows Script Host) script
Specify the WSH script execution program and the path of the WSH file.
E.g.) remote+ssh://CScript.exe //nologo C:\PerformanceMonitor\Script\test.vbs
• PowerShell script
Specify the PowerShell script execution program, the path of the Power Shell script, and the "<
NUL" string that corresponds to the operation of pressing the enter key (< NUL).
E.g.) remote+ssh://PowerShell.exe -Command "C:\PerformanceMonitor\Script\test.ps1" <
NUL
For Linux
Specify the Linux shell execution program and the path of the Linux shell script.
E.g.) remote+ssh://sh /PerformanceMonitor/test.sh
System Monitor - Performance Monitoring Services gets the performance data and error information from
the result of script execution as the information of the standard output. So, the script must output the result
to the standard output in the following formats:
When the script execution succeeded
Output the result to the standard output in the "@metric number [name=Performance Indicator]"
format. You can write a remote user script so that it outputs the performance data for multiple
performance indicators to a same script. In this case, allow the script to output multiple lines of
@metric.
The number above indicates performance data. When specifying number, use integers, fractions, and
floats.
The name outputs the title for a custom performance indicator defined in System Monitor -
Performance Monitoring Services.
If you make the script output several lines of @metric, you must specify name to be output to the
execution result to identify the performance indicator.
15
If the script outputs only one "@metric" line, you can omit to specify name. In this case, the output is
collected as the data of the performance indicator which sets the script to be executed.
For datetime, specify the time when the performance data is collected. System Monitor - Performance
Monitoring Services handles the datetime output as the data collection time.
datetime is optional. If it is omitted, the time when the script was executed by System Monitor -
Performance Monitoring Services is considered as the data collection time.
For the format in which datetime is specified, see the description later.
The following shows an output example:
echo @metric 1.1E+2 name=CustomPerformance datetime="2014/01/01 09:00:00"
When the script execution failed
Output the result as the error code to the standard output in the "@metric [name=Performance
Indicator] [error=Error Message] [errorcode=Error Code]" format. Only integers are available for a
numeric part of the error code. The Error and Error Code are optional.
Provide the script with the error, which indicates the error message when the script fails, to output it.
Output the errorcode, which is an arbitrary code to categorize errors in a script, in an integer.
For example, specify "echo @metric name=CustomPerfromance error=Failed to execute the
command errorcode= -1" to output the error code.
If the script outputs multiple "@metric" lines, you must specify name to be output to the execution
result to identify the performance indicator.
If the script outputs only one "@metric" line, you can omit to specify the value "name." In this case, the
output is handled as the error information of the custom indicator which sets the script to be executed.
For the Windows monitored machine, use the default encoding to create a script file. Avoid other text
strings than the default encoding to be contained in the script.
For the Linux monitored machine, use the ASCII encoding or the UTF-8 (without BOM) encoding to
create a script file.
Local user script
You can create custom performance indicators by specifying a script on the management server. When
collecting performance data, System Monitor - Performance Monitoring Services executes the script on
the management server at every collecting interval. Then, System Monitor - Performance Monitoring
Services extracts the results from the [standard output] information to obtain the performance data.
When creating a custom performance indicator, you must specify the name of the script file on the
management server as the executing string. You must add the string "local://" before the executing
string. Use the following execution examples by script type as a reference. We recommend that you use
the full path to specify the script path to avoid execution errors.
You can specify scripts such as batch files, Windows Script Host (WSH) and PowerShell on the
management server.
16
• Batch file
Specify the batch file path.
E.g.) local://C:\PerformanceMonitor\Script\test.bat
• WSH(Windows Script Host) script
Specify the WSH script execution program and the path of the WSH file.
E.g.) local://CScript.exe //nologo C:\PerformanceMonitor\Script\test.vbs
• PowerShell script
Specify the PowerShell script execution program, the path of the Power Shell script, and the "<
NUL" string that corresponds to the operation of pressing the enter key (< NUL).
E.g.) local://PowerShell.exe -Command "C:\PerformanceMonitor\Script\test.ps1" < NUL
System Monitor - Performance Monitoring Services obtains the performance data and error information
from the result of script execution as the information of the standard output. The script must output the
result to the standard output in the following formats:
When the script execution succeeded
Output the result to the standard output in the "@metric number name=Performance Indicator
target= Target Machine [pathtype=pvm|sysmon] [datetime=Collection time]" format. You can write
a local user script so that it outputs the performance data for multiple monitored machines and
performance indicators to a same script. You must specify so that it outputs both name and target to
identify the performance indicator and the monitored machine.
The number above indicates performance data. When specifying number, use integers, fractions,
and floats.
The name outputs the title for the custom performance indicator which is defined in System Monitor
- Performance Monitoring Services.
The target indicates the path of System Monitor - Performance Monitoring Services or
SigmaSystemCenter which is required to identify the target machine. You can use the pathtype to
explicitly specify whether the path specified in target is on System Monitor - Performance
Monitoring Services (sysmon) or on SigmaSystemCenter (pvm).
You need to provide the target output with a full path to the monitored machine on System Monitor -
Performance Monitoring Services or SigmaSystemCenter in the format Group Name\Machine
Name. If pathtime is omitted, the path specified in target is considered it is either on System
Monitor - Performance Monitoring Services or on SigmaSystemCenter. However, if the path is on
both, it is considered the path on System Monitor - Performance Monitoring Services.
The datetime is the data collection time specified by a script output. System Monitor - Performance
Monitoring Services considers datetime is the time when the performance data was collected.
datetime is optional. If it is omitted, the time when the script was executed by System Monitor -
Performance Monitoring Services is considered as the data collection time.
For the format in which datetime is specified, see the description later.
17
The following shows an output example:
echo @metric 1.1E+2 name=CustomPerformance target=Group1\Machine1 pathtype=pvm
datetime="2014/01/01 09:00:00"
When the script execution failed
Output the result as the error code to the standard output in the format "@metric
name=Performance Indicator target=Target Machine [error=Error Message] [errorcode=Error
Code]". You do not have to specify error or errorcode because they are optional.
Provide the script with the error, which indicates the error message when the script fails, to output
it. Output the errorcode, which is an arbitrary code to categorize errors in a script, in an integer.
For example, specify "echo @metric name=CustomPerfromance error=Failed to execute command
errorcode= -1" to output the error code.
You must specify so that it outputs both name and target to identify the performance indicator and
the monitored machine on which an error occurred.
Rules for specifying options
Use space characters to separate options.
If the specified text string of an option contains space characters, enclose the string with double
quotation marks (").
If the specified text string of an option contains equal signs (=) and double quotation marks ("), add
a backslash symbol (\) before an equal sign or double quotation mark to escape them.
<Example>
When the performance indicators titled "customperf" on a monitored machine, whose group name is
"Group 1" and whose machine name is "my=mach"ine ", specify the option as follows:
echo @metric 50 name=customperf target="Group 1\my\=mach\"ine"
About the datetime format
Specify datetime in the format "yyyy/MM/dd HH:mm:ss". If you set the different time zone to a
monitored machine from that of the System Monitor - Performance Monitoring Services management
server, provide the strings to specify the time zone in the format "yyyy/MM/dd HH:mm:sszzz".
yyyy: Provide the year.
MM: Provide the month.
dd: Provide the day.
HH: Provide the hour in a 24-hour time format.
mm: Provide the minute
ss: Provide the second.
18
zzz: Provide the difference between the time in the time zone and Coordinated Universal Time
(UTC); for example, provide "+09:00" for the Japan time zone.
If omitted, the setting of the local time on a management server is set to this option.
When you intend to output multiple performance data with a single user script
If you intend to execute multiple remote user scripts for the same monitored machine or if you intend to
execute multiple local user scripts on a management server, we recommend that you write a user script
so that it outputs the performance data for multiple monitored machines and performance indicators
because you can take the following advantages:
• You only have to specify a same user script for multiple custom performance indicators. This
reduces the number of user scripts and makes it easier to manage user scripts.
• This can reduce the number of times user scripts execute. If a user script has already executed at
the same timing of collecting data, the data which has been output at the time is obtained as a
performance data without executing the user script.
System Monitor - Performance Monitoring Services can cache the multiple performance data if
they have been output in a same script, which makes it possible to reduce the number of times of
the script execution as mentioned above.
1.3.4 Storing Data that has been Collected
System Monitor - Performance Monitoring Services automatically delete performance data after a certain
amount of time has passed. The storage period can be set between one day and 30000 days. The default value is
3 days. You can change the storage period using the database management tool of System Monitor -
Performance Monitoring Services. For how to change the storage period, see, 7.1 "Data Management Tool."
Although the default storage period is three days, the performance data can be preserved for longer period
because they are aggregated at various collecting intervals.
Use the script that can be executed in a short time wherever possible to
prevent it from affecting on the behavior of collecting performance indicators.
Note that long-running scripts may cause execution delays or failures on
other scripts.
With multiple scripts, System Monitor - Performance Monitoring Services
sequentially executes those scripts and a script waits until that the other
script execution has completed if other script is running. If the long-running
script exists within them, it may cause an execution delay on other scripts.
Also, the script execution error may be caused by the script whose execution
exceeds fixed timeout duration (30 seconds).
Note
19
1.4 Summary Data
System Monitor - Performance Monitoring Services manages data collected from the machine being monitored
as well as summary data obtained through statistical analysis of the collected performance data. Summary
data, which is obtained through statistical analysis of data collected over a certain period of time (referred to
below as the “summary interval”), is used to accelerate graph display of data collected over a long period and to
store performance data efficiently.
1.4.1 Storage of Summary Data
System Monitor - Performance Monitoring Services manages summary data collected at four summary intervals.
If the summary interval is small, performance data of the system can be stored in detail, but the amount of data
becomes large. If the summary interval is large, performance data cannot be saved in detail, but performance
data can be stored for an extended period of time. The storage period of data collected at each summary interval
can be changed according to the usage of historical performance data or the free space of the management
server.
The default storage periods for the summary data are as follows:
Summary
interval
Storage period
5 minutes 1 week
15 minutes 1 month
1 hour 3 months
1 day 5 years
To change the storage period for the data, use the System Monitor Data Management Tool. For details on the
Data Management Tool, see "7.1 Data Management Tool".
1.4.2 Usage of Summary Data
System Monitor - Performance Monitoring Services uses summary data for displaying graph. The data to be
used depends on the plot interval of a graph.
Plot Interval Available Data
less than 5 minutes collected data
5 minutes to less than 15 minutes 5-minute summary data
15 minutes to less than 1 hour 15-minute summary data
1 hour to 1 day or less 1-hour summary data
more than 1 day 1-day summary data
20
1.5 Data Management
1.5.1 Data Management Tool
Data Management Tool can be used to change the storage period for the performance data and summary data or
to summarize the data. The Data Management Tool can be used when:
• The storage period is to be extended to store performance data over a longer period.
• The storage period is to be reduced to save data storage space.
• Data summarization is performed to utilize the data used in the previous version as summary
data.
For details on the Data Management Tool, see "7.1 Data Management Tool".
1.5.2 Database Size Monitoring Function
System Monitor - Performance Monitoring Services saves the performance data collected from the machine to be
monitored, as well as summary data that is obtained through summarization of the performance data in a
database. The database size monitoring function outputs a warning log to notify the user when the database
size exceeds the specified warning value. This function allows the user to be aware of the database engine
exceeding a given size, or the lack of free disk space on the management server.
21
1.6 Groups
System Monitor - Performance Monitoring Services can organize machines into groups and then monitor and
graph the performance status of each group. This makes it possible to track the performance status trends for
each group (for example, a group of machines with the same task) and for the entire system. When a
performance problem (such as high load) occurs, it is possible to work out which machine is the cause of the
problem by switching from group-level view to machine-level view.
System Monitor - Performance Monitoring Services expresses grouping units as nodes. There are three types of
nodes: the management server node (management server), group nodes (groups), and machine nodes (machines).
Machine nodes represent individual machines that have not been grouped.
Groups can be created out of any combination of machines, except that each machine can only belong to one
group. If configuration is to be synchronized with SystemProvisioning, then the approach to grouping machines
depends on SystemProvisioning. See "1.10 The SystemProvisioning Linkage Function" for information on
applying SystemProvisioning configuration information.
22
Although System Monitor - Performance Monitoring Services supports both
IPv4 addresses and IPv6 addresses, the available IP address protocol
depends on whether the operating system on the monitored machine
supports the protocol. In all operating systems except Windows and Linux,
only IPv4 address is available because the access using IPv6 addresses to the
monitored machine running those operating systems is not supported. When
you use only a machine name to access the managed machine which can be
accessed with IPv4 addresses only, configure the machine name to resolve to
the IPv4 address.
1.7 Connecting to Monitored Machines
The following section provides general notes about the connection settings for the monitored machines.
To collect performance data for "Standard Performance Indicator", specify IP address, machine name and access
account of the monitored machine on System Monitor - Performance Monitoring Services because the monitored
machine is directly accessed for collecting performance data. For collecting performance data only for
"Performance Indicator for Virtual Machine" or "Performance Indicator for Physical Machine", the note shown
below is not necessary to be considered because these settings need to be applied from SystemProvisioning and
the monitored machine is not directly accessed for collecting performance data. For the type of performance
indicator, see "1.3 Collected Performance Data".
1.7.1 Specifications for Monitored Machines
Specify the IP Address (Optional) and the Machine Name (required) of the monitored machine. These are used
according to the following rules.
(1) When the IP address is set, System Monitor - Performance Monitoring Services accesses the monitored
machines using the specified IP address.
(2) When the IP address is not set, System Monitor - Performance Monitoring Services accesses the monitored
machines using the machine name.
When an IP address of the monitored machines is not specified using the System Monitor - Performance
Monitoring Services Management Console, the machine name is used to access the monitored machine. The
machine name is resolved to an IP address by the Management Server automatically. An access error results
when it cannot be converted to an IP address. To enable name resolution, you must register the machine name
in DNS or the management server hosts file. Specifically, when the monitored machine OS is Linux, VMware
ESX/ESXi, Citrix XenServer or KVM, the machine name is not resolved by default.
See “Adding Machines” and “Adding Groups” in “4.1.1 Specifications in the Navigation Tree Window” for
specifying machine names and IP addresses with System Monitor - Performance Monitoring Services.
Note
23
When Citrix XenServer is added as a monitored machine in the navigation
tree window, confirm that the machine is a pool master machine. If the
machine is a pool member machine, the performance data of the machine
cannot be collected. When a pool member machine is added as a monitored
machine, use the SystemProvisioning linkage function. See "1.10 The
SystemProvisioning Linkage Function" for more detail.
The SSH server software installed on a Windows monitored machine must
support IPv6 to use IPv6 addresses in collecting performance data with
scripts on a Windows machine.
1.7.2 Connection Setting for the Monitored Machines
Specify the account name and password so that the Performance Monitoring Service can access the monitored
machine. When using a SSH public key authentication method, use SECSH format for the file format of the
private key file located in the management server. Use OpenSSH format for the file format of the public key
files located in the monitored machines. When VMware ESX/ESXi is the monitored machine, VMware Web
Service is used to access the monitored machine, and protocol setting is ignored. When Citrix XenServer is the
monitored machine, XML-RPC protocol is used to access the monitored machine, and protocol setting is ignored.
Connection settings can be specified for each management server, each group or each individual machine.
• For each management server
All machines under the management server are applicable targets. Use the Connection tab in the Configuration
window.
• For each group
All machines under the group are applicable targets. When connection settings are different for each group, you
must specify connection settings for each group. Use the Connection tab in the Group Settings window.
• For each machine
Applicable only to the machine. When connection settings are different for each machine, you must specify
connection settings for each machine. Use the Connection tab in the Machine Settings window.
Note
Note
24
Targets Setting window Reference
For each management
server
All machines under the
management server
The Connection tab in the
Configuration window.
“2.3.1 Setting the
Connection Account to
the Monitored Machine”
For each group All machines under the
group
The Connection tab in the
Group Settings window
“4.1.1 Specifications in
the Navigation Tree
Window”
For each machine Only the machine The Connection tab in the
Machine Settings window
“4.1.1 Specifications in
the Navigation Tree
Window”
The following figure shows which setting will be valid when you specify connection settings for management
server, group and machine.
The setting for the management server is valid
The setting for the machine is valid
The setting for the group is valid
Specified for the machine
Specified for the group
YES
YES
NO (Default)
NO (Default)
25
1.7.3 Settings on the Monitored Machine
You must make the following settings on the monitored machine for System Monitor - Performance Monitoring
Services to connect to the monitored machines.
Windows
• For monitored machines running Windows, the following settings must be made on the monitored machines:
♦ Set the account name and password as specified with connection settings. The connection account must
belong to the Administrators group or Performance Monitor Users group of the monitored machine.
User privilege
Administrators
Group
Performance
Monitor Users
Group
Target OS Type Windows XP x -
Windows Vista x x
Windows 7 x x
Windows 8 x x
Windows 2000 Server x -
Windows Server 2003 x x
Windows Server 2008 x x
Windows Server 2012 x x
♦ Configure the Security Settings for the account and for the group which that the account belongs to as
follows:
• Assign the "Access this computer from the network" right to the account or to the Administrators group
or the Performance Monitor Users group which the account belongs to.
• Avoid that the account and the Administrators group or the Performance Monitor Users group which
the account belongs to are added to the list of the "Deny access to this computer from the network" right.
♦ Local Security Policy: When the monitored machine OS is Windows Vista, Windows 7, Windows Server
2008, Windows 8 or Windows Server 2012 and an Administrators group account is used for the access
account for a target machine, perform the following steps. This setting is not necessary if a domain user is
used.
1. Start the Local Security Policy using the Administrative Tools in Control Panel
2. The Local Security Settings are appeared. Select Security Options of Local Policies from the tree on
the left side.
3. When the built-in Administrator is used;
Display the properties by double-clicking User Account Control: Admin Approval Mode for the
Built-in Administrator account.
26
When an Administrators group account other than the built-in Administrator is used;
Display the properties by double-clicking User Account Control: Run all administrators in Admin
Approval Mode.
4. Select Disabled and then Click OK.
♦ Local Security Policy: When the monitored machine OS is Windows XP, perform the following steps.
1. Start the Local Security Policy using the Administrative Tools in Control Panel
2. The Local Security Settings are appeared. Select Security Options of Local Policies from the tree on
the left side.
3. Display the properties by double-clicking Network access: Sharing and security model for local
accounts.
4. Select Classic – local users authenticate as themselves and then Click OK.
♦ Service: Start the following services on the monitored machines.
• Remote Registry
• Server
Perform the following steps.
1. Start Services using Administrative Tools in the Control Panel.
2. A list of the services is appeared. Select the service display name described above and then double-click
to open.
3. Set Startup type in General tab to Automatic.
♦ Firewall: When the Windows Firewall on the monitored machine is enabled, set the exceptions for Windows
Firewall using the following steps.
1. Start Windows Firewall in the Control Panel.
2. Select the Exceptions tab in the Windows Firewall window.
3. Check File and Printer Sharing.
If Windows 7, Windows Server 2008, Windows 8 or Windows Server 2012 is the monitored machine, enable
the rules shown below on Windows Firewall settings.
• File and Printer Sharing (NB-Name-In)
• File and Printer Sharing (NB-Session-In)
• File and Printer Sharing (SMB-In)
♦ Power Options: Set not to go to sleep mode or standby mode automatically. If a monitored machine goes to
these modes, the performance data from this machine may not be collected.
27
• The following settings are required to collect performance data with the remote user script.
♦ In the scripted performance data collecting, the script specified in Performance Indicator Setting is
executed after connected to a monitored machine from a management server using SSH, which requires
installation of the SSH server software on the Windows monitored machine. The validated SSH server
software includes OpenSSH (cygwin) 6.1 or FresSSHD 1.2.6.
♦ In Connection Setting, specify the account and password for the SSH connection. The account must belong
to the administrators group on the monitored machine.
♦ To connect to the Windows monitored machine, specify "cmd.exe" as the login shell when connecting. In the
SSH server configuration with OpenSSH (cygwin), the default login shell for cygwin is "bash", so, change
the login shell to "cmd.exe". The login shell for the account configured with /etc/passwd is in the installation
folder of cygwin.
♦ Windows PowerShell must be installed on a Windows monitored machine to collect performance data using
PowerShell, and the user you want to use must be permitted to execute PowerShell remotely. To do so,
perform the followings.
1. Log on to the monitored machine with the account that you want to use.
2. Click Start, click Run, and then type "PowerShell.exe" to start PowerShell.
3. Execute "Set-ExecutionPolicy RemoteSigned -Scope CurrentUser" to grant the execution right for
PowerShell to remotely execute PowerShell to the account.
♦ Firewall:When a firewall has been setup, exempt SSH communications from firewall blocking.
♦ Enable password authentication if you want to use the password authentication scheme. If you want to use
public key cryptography, prepare the public key file (the OpenSSH format) in advance.
• In collecting the performance data using a local user script, configure the management server so that the
script can be executed on the management server as follows:
♦ Collecting the performance data by using PowerShell requires the right to execute PowerShell scripts on
the management server. To allow the script to be executed on a management server, perform the following:
1. Log on to the management server with an account that has administrator rights.
2. Click Run from the Start menu, and then enter "PowerShell.exe" to start PowerShell.
3. Run "Set-ExecutionPolicy RemoteSigned" to grant the right to execute PowerShell on a management
server.
Linux/KVM
• For monitored machines that are running Linux or KVM is the monitored machine, the following settings
must be made on the monitored machines:
♦ Set the account name and password as specified with connection settings.
♦ Enable SSH on each monitored machine. When a firewall has been setup, exempt SSH communications
from firewall blocking. If a public key authentification is used, then prepare the public key on the
monitored machine.
28
♦ Set "C" to the environment variables "LC_ALL" and "LANG" of the account.
VMware ESX/ESXi
• If VMware ESX/ESXi is the monitored machine, the following settings must be made on the monitored
machines:
♦ Set the account name and password as specified with connection settings.
♦ Assign a role other than "No Access" to the connection account using Virtual Infrastructure Client.
♦ VMware Web Service is used to access the monitored machine. When a firewall has been set up, exempt
SSL communications from firewall blocking.
♦ VMware Web Service is used to access the monitored machine, and protocol setting is ignored.
Citrix XenServer
• If Citrix XenServer is the monitored machine, the following settings must be made on the monitored
machines:
♦ Set the account name and password as specified with connection settings.
♦ When a firewall has been set up, exempt SSL communications from firewall blocking.
♦ XML-RPC protocol is used to access the monitored machine, and protocol setting is ignored.
1.7.4 Ports Used between Management Servers and Monitored Machines
The following ports are used for communications between the management server and monitored machines.
Management Servers Protocol Monitored Machines Explanation
assigned
automatically
TCP NetBIOS over TCP/IP
(139)
Direct Hosting SMB
Service (445)
Used when Windows performance
data is collected. *1
UDP NetBIOS over TCP/IP
(137)
assigned
automatically
TCP SSH (22) *2 Used when Linux/KVM/Windows
performance data is collected using
SSH.
assigned
automatically
TCP SSL (443) Used when Citrix XenServer/VMware
ESX/ESXi performance data is
collected. *1 If one of the NetBIOS (UDP-137, TCP-139) or SMB/CIFS (TCP-445) setting is enabled, the
Windows performance data can be collected.
*2 In Windows monitored machine, the SSH (22) needs to be opened only when the performance data will be collected by executing a remote user script.
29
1.8 Displaying Performance Status and CSV File Output
System Monitor - Performance Monitoring Services displays performance data for the system graphically.
Displaying performance data graphically enables trends to be identified intuitively, and makes it easier to
analyze the operational status of the system.
Graphs of performance status information can be displayed for each group or for each individual machine.
Performance data is displayed as statistically calculated values. You can also output performance data in the
graph display on the management console to a CSV file or a TEXT file.
In addition to the feature of the management console shown above, there is a command line feature to output
performance data.
1.8.1 Statistics
"Statistics" are the statistical methods that are used to calculate the performance data collected during the plot
interval when performance data is graphed.
• Maximum value: Shows the maximum value.
• Average value + standard deviation: Shows the relationship between the average value and the degree of
scattering for all data.
• Average value: Shows the average value.
• Average value - standard deviation: Shows the relationship between the average value and the degree of
scattering for all data.
• Minimum value: Shows the minimum value.
• Weighted average value: Shows the average value when weights that have been set up for
monitored machines are applied to the data that has been collected.
This method is effective when checking the performance status of
groups that are made up of machines with different specifications.
• Summation value: Shows the summation value of all machines that belonged to the selected
Group or Root node. When a machine node is selected, it shows the same
value as the Average value.
Weights can be specified for each monitored machine, from 0.0 to 10000.0. For example, suppose there is a
group that includes machines with different numbers of CPUs. If these machines are compared using the same
performance index, machines with more CPUs are displayed as having lower usage rates even if the work load
is the same. In this kind of situation, the processing ability of the group can be expressed more accurately by
weighting machines according to the number of CPUs that they have. See Appendix A, "How performance
values for groups are calculated" for guidelines on weightings.
30
1.8.2 Displaying Formats for Performance Status Graphs
System Monitor - Performance Monitoring Services displays Performance Indicators using three different
graph formats, as described below.
(1) Node Comparison Display
In this format, the performance status for a particular Performance Indicator item and a particular
statistic are displayed for each node. This format is valid for situations such as comparing groups or
checking whether there is anything unusual about the performance status of a particular machine in a
certain group, in terms of, for example, the average values for CPU usage.
Figure 1-1: Displaying the Performance Status of Each Group
31
Figure 1-2: Displaying the Performance Status of a Group and a Particular Machine within the Group
(2) Performance Indicators Display for a Specified Node
With this format, the performance status for a particular node and a particular statistic is displayed for
each Performance Indicator item. This format is valid for situations such as investigating performance
bottlenecks for a group by comparing, for example, the average values for CPU usage with the average
values for disk transfer rates. Individual machines can be handled in the same way as groups.
Figure 1-3: Display Comparing CPU Usage and Disk Transfer Rate
32
(3) Statistical Display for a Specified Node and Performance Indicators
With this format, the performance status for a particular node and a particular Performance Indicator item
is displayed for each statistic. This format is valid for situations such as checking, for example, changes in
the maximum, average and minimum values of CPU usage for a group. Individual machines can be handled
in the same way as groups.
Figure 1-4: Displaying CPU Usage
To display the performance status for a particular group or machine for a particular Performance Indicator
item and a particular statistic (that is, to select particular elements for each factor), any of the display
methods (1) to (3) above can be used.
1.8.3 Graph Display Periods
System Monitor - Performance Monitoring Services graphical displays can display performance status
information in real time or historical information for data that has been collected in the past.
(1) Realtime Display
Display the latest performance status information for a specified display period from the present time. The
graph is updated according to specified update interval. This kind of display is suitable for situations where
the current performance status needs to be checked in real time.
(2) Historical Display
Display the performance status for a specified period from a specified starting time. The graph is not
updated periodically. This kind of display is suitable for situations where the past performance status
needs to be checked.
33
With real-time displays, if performance data for the server being displayed
cannot be collected within the collecting interval, data will not be displayed
on the graph for the section where data has not been collected. The same
applies if the performance status is displayed for a management server or a
server group that includes a server for which performance data could not be
collected.
If performance data cannot be collected within the collecting interval three
times in a row, the server is regarded as having error status, and the graph
will not be displayed until the error status is recovered. For graphs of server
groups that include a server with error status, the performance status for the
remaining normal servers is displayed.
Note
Note
34
1.8.4 The Graph Window
The System Monitor - Performance Monitoring Services management console GUI can arrange and display
multiple graphs. Multiple node comparison performance status graphs can be arranged and displayed.
Alternatively, node comparison performance status graphs can be arranged and displayed together with
Performance Indicator comparison status graphs. This is effective when there is a need to keep track of the
status of the system from multiple perspectives.
Figure 1-5: Displaying Multiple Windows
35
For Windows XP and Windows Server 2003, set with Language of
non-Unicode programs in the Advanced tab of Regional and Language
Options from Control Panel. Normally, ANSI is used for English and
Shift-JIS is used for Japanese.
1.8.5 Output Graph Data to File
You can output performance data displayed in the graph window to an external file such as CSV with System
Monitor - Performance Monitoring Services.
Files are output in the following formats.
Line 1 Summary
Line 2 “Time” Column Name Column Name …
Line 3 Time Data Value Data Value …
… Time Data Value Data Value …
… … … … …
File summary is output to Line 1, performance data explanatory headings are output to Line 2, and time
(Column 1) and performance data (from Column 2 onwards) are output from Line 3 onwards. Data start time,
data end time, data time period, target node name, performance indicator title and statistical method name are
output for summaries. For column names, type names of a comparative perspective are output according to
the graph type specified from among node names, performance data names and statistical method names as
names identifying performance data of identical rows from Row 3 onwards. For time, time axis values of each
point plotted on the graph are output. Vertical axis values of points plotted on the displayed graph are output
for data values.
The column separator and encode format of output characters are as shown below according to the file type
specified at file output.
Specified File Type Column Separator Encode Format of Output
Characters
CSV (Separated by Comma) Comma System default (current) encode
Text (Separated by Tab) Tab System default (current) encode
UTF-8 Text (Separated by
Tab)
Tab UTF-8
1.8.6 Output Data Using Command Line Tool
The ssc-perf command can display the performance data collected by System Monitor - Performance Monitoring
Services on the command prompt. See "8.1 ssc-perf Command" about the ssc-perf command.
Note
36
1.9 Threshold Monitoring and the Notification Function
System Monitor - Performance Monitoring Services can detect and report load status errors with monitored
machines by monitoring thresholds for the performance data that is collected. Performance fault events can also
be notified to SigmaSystemCenter.
Thresholds can be set up for Performance Indicators that are currently being collected, both for groups and for
individual machines.
1.9.1 Types of Thresholds
The following four types of thresholds can be set up for a single item of Performance Indicator that is subject to
threshold monitoring. More than one type of threshold can be selected.
• Upper-limit critical value: The status is regarded as "critical" if the target performance data
reaches or exceeds this value.
• Upper-limit warning value: The status is regarded as "warning" if the target performance data
reaches or exceeds this value.
• Lower-limit critical value: The status is regarded as "critical" if the target performance data drops
to or below this value.
• Lower-limit warning value: The status is regarded as "warning" if the target performance data
drops to or below this value.
These values must meet the following relationship:
(Lower-limit critical value) < (Lower-limit warning value) < (Upper-limit warning value) < (Upper-limit critical
value)
1.9.2 Units of Threshold Monitoring
Thresholds can be set up for the following units.
Monitoring unit Explanation
Machine Threshold monitoring is performed for the specified machine.
Group Threshold monitoring is performed for the specified group. Performance data for the group (produced by statistically processing the performance data for all of the machines in the group) is used as the monitoring target.
All machines in group
Threshold monitoring is performed for each machine in the specified group. The Performance Indicator that is subject to threshold monitoring is the same for all machines in the group, but the threshold monitor monitors the performance data for each machine separately.
1.9.3 Performance Indicators Subject to Threshold Monitoring
Performance Indicators that are currently being collected can be set to be subject to threshold monitoring. If the
monitoring unit is "group", the statistic must also be specified.
37
1.9.4 Notification Actions at Over-Threshold
The following actions can be performed at over-threshold:
Write an event to the event log: This is the default behavior. This cannot be changed.
Display in the navigation tree: Uses an icon to display the status of the machine or group in the
over-threshold status. This is the default behavior. This cannot be
changed.
Display in the Log window: This is the default behavior. This cannot be changed.
Notify SystemProvisioning: SystemProvisioning can be notified at over-threshold and at recovery. For
notification, set notification category to notice. Set actions for notification
category using the SigmaSystemCenter policy. See "1.10.6 Notifying
Performance Abnormalities" for information on notification categories.
In order to distinguish between temporary load states and ongoing load states, it is also possible to configure
the extent or number of over-threshold alert until the load state is considered "critical". The parameters specify
the number of times the load state is checked. You can configure the parameters to define how many instances
of over-thresholds before an abnormality is reported. By default, an abnormality is reported if over-threshold is
found ten times out of ten checks. The check interval is the same as the data collecting interval.
You can also configure the amount of time before another report is made if the over-threshold status has not
improved. By default, another report is made every 30 checks.
Example 1: Judging Over-Threshold Status
This example shows how over-threshold status is judged when over-threshold status has been set as four
threshold crossings out of five checks for upper-limit threshold monitoring. Data is ordered in time from left to
right, so that the most recent data is at the far right. In this example, at the point when the data at the far right
is collected, threshold crossings have been verified for four of the last five checks, and so the status is judged to
be "over-threshold" and the action that has been set up by the user is executed.
4/5
Status judged as "over-threshold"
Threshold Data over the threshold Data under the threshold
1/5 1/5 1/5 1/5 2/5 3/5 2/5 2/5 Count:
38
Example 2: Judging Recovery from Over-Threshold Status
This example shows how recovery from over-threshold status is judged when over-threshold status has been set
as four threshold crossings out of five checks for upper-limit threshold monitoring. In the same way as for when
over-threshold status is judged, at the point when the data at the far right is acquired, performance data has
been verified as being under the threshold for four of the last five checks, and so the status is judged to have
recovered from over-threshold status. Threshold checks for determining recovery from over-threshold status
start counting from the next performance data after the data for which over-threshold status was decided.
4/5
Status judged as having recovered from over-threshold status
0/5 0/5 1/5 1/5 2/5 3/5 2/5 3/5
Threshold Data over the threshold Data under the threshold
Count:
Example 3: Judging that Over-Threshold Status Continues
The specified notification action can be executed again if recovery from over-threshold status cannot be
recognized within the specified period (number of checks) from when the over-threshold status was first
verified.
Monitoring for over-threshold status
Monitoring for recovery from over-threshold status
Notification Repeat notification
If recovery is not recognized within the specified number of checks since the last notification, a notification with the same content will be made again.
Data over the threshold Data under the threshold
39
1.10 The SystemProvisioning Linkage Function
System Monitor - Performance Monitoring Services has functions that applies system configuration information
from SystemProvisioning and notifies SystemProvisioning of performance abnormalities as one of the
SigmaSystemCenter components.
1.10.1 System Configuration
The system configuration for System Monitor - Performance Monitoring Services and SystemProvisioning is as
follows. During the applying process for system configuration information, the System Monitor - Performance
Monitoring Services performance monitoring service acquires configuration information from
SystemProvisioning and applies this to the configuration information for System Monitor - Performance
Monitoring Services. When performance abnormalities occur, an event is notified to SystemProvisioning. The
Performance Monitoring Service and SystemProvisioning do not have to be on the same management server.
Management server
System Monitor - Performance
Monitoring Services Management
console
Collecting performance data
Saving performance data
System Monitor - Performance Monitoring Services Performance Monitoring Service
Displaying graphs
Making settings
Monitored machines
SigmaSystemCenter(SystemProvisioning)
Notifying performance abnormalities
Applying system configurations
Displaying graphs
Making settings
SigmaSystemCenter Web console
40
1.10.2 Applying System Configuration Information
SigmaSystemCenter configuration information can be applied to System Monitor - Performance Monitoring
Services. For applying SystemProvisioning configuration information to System Monitor - Performance
Monitoring Services, it is necessary to set mapping settings of their configuration information. There are two
ways to set mapping settings. One is to enable the performance monitoring setting on SigmaSystemCenter; the
other is to enable applying of the configuration information on System Monitor - Performance Monitoring
Services.
(1) Method by enabling the performance monitoring setting for groups or models of SigmaSystemCenter
Configure to apply the configuration information to System Monitor - Performance Monitoring Services in
the monitoring settings on the group/model property of the SigmaSystemCenter Web console. At the timing
of applying configuration information of SystemProvisioning, the corresponding group, whose "Performance
Data Collection Settings" has been enabled on the property of SigmaSystemCenter group/model settings, is
added automatically on System Monitor - Performance Monitoring Services and the configuration
information of the managed machines in the specified SigmaSystemCenter group or model shown below is
applied to the configuration information of System Monitor - Performance Monitoring Services.
・ Machine name
・ IP address
・ OS information
・ Machine status (normal or error status)
・ Account information
・ Collecting Data settings (Monitoring Profile)
・ Threshold Monitor settings (Monitoring Profile)
When the performance data collecting setting is enabled in the group property of SigmaSystemCenter, if the
group has several models, multiple groups mapped to their corresponding models will be created on System
Monitor - Performance Monitoring Services. If a group has a model in itself and if collecting performance
data is enabled for the higher group property or the target model property, the group for the model is created
on System Monitor - Performance Monitoring Services.
The group created on System Monitor - Performance Monitoring Services will be named like "Category name
- Group name" if the target to be applied is a group, "Category name - Group name - Model name" if the
target to be applied is a model.
System Monitor - Performance Monitoring Services acquires the account settings and the monitoring profile
settings (include collecting data settings and threshold monitor settings) from the Performance Monitor
settings of the group or model property of SigmaSystemCenter and applies automatically to its own group
settings. The collecting data settings and threshold monitor settings applied from SystemProvisioning
cannot be specified flexibly because the monitoring profile is not editable with System Monitor - Performance
Monitoring Services. But collecting performance data and threshold monitoring on System Monitor -
Performance Monitoring Services can be started more easily and automatically without using System
41
Monitor Management Console when SystemProvisioning and System Monitor - Performance Monitoring
Services are configured on the same management server. For the default settings of the monitoring profile,
see "Appendix" of SigmaSystemCenter Configuration Guide.
A collecting data setting and threshold monitor setting specified in a monitoring profile is not editable via
the System Monitor Management Console. But via SigmaSystemCenter Web Console or with ssc command,
it can be done. See SigmaSystemCenter Web Console Reference Guide for details on Web Console and see
SigmaSystemCenter ssc Command Reference for details on ssc command. It is possible to configure
collecting data settings other than the settings specified in the monitoring profile with the Collecting Data
Setting List dialog box of System Monitor Management Console, and possible to configure threshold monitor
settings with the Set Threshold Monitor dialog box.
An access account can be configured on the Performance Monitor settings of SigmaSystemCenter and on the
Connection settings of System Monitor - Performance Monitoring Service. You can select which account is
used as an access account via the System Monitor Management Console. See "4.1 Specifications for
Monitored Machines" for details on how to configure this setting.
(2) Method with group settings of System Monitor - Performance Monitoring Services
Enable Apply SystemProvisioning Configuration for the group in System Monitor - Performance
Monitoring Services, and specify the corresponding SystemProvisioning configuration full path to the
group/model on the group settings of System Monitor - Performance Monitoring Services. At the timing of
applying configuration information of SystemProvisioning, the configuration information of the managed
machines in the specified SigmaSystemCenter group or model shown below is applied to the configuration
information of System Monitor - Performance Monitoring Services.
・ Machine name
・ IP address
・ OS information
・ Machine status (normal or error status)
Account settings and collecting data settings have to be specified on System Monitor - Performance
Monitoring Services. Because the collecting data settings are specified on the Collecting Data Setting List
dialog box of System Monitor Management Console, finer settings for each node are possible. For the
Collecting Data Setting List dialog box, see "4.2.2 How to Specify Collecting Data Settings".
System Monitor - Performance Monitoring Services uses the Management IP address already set up using
SystemProvisioning to connect to the target machines. Specify the Management IP address in the Host Setting
window with the SigmaSystemCenter Web Console. When the Management IP address is not specified, System
The machine only in the immediate child group of SigmaSystemCenter can
be registered as a monitored machine to the group of System Monitor -
Performance Monitoring Services which corresponds to the group of
SigmaSystemCenter.
Note
42
When you specify a machine name, IP address has to be resolved as an IPv4
address for the monitored machine running other operating system than
Windows and Linux.
Monitor - Performance Monitoring Services uses the machine name to connect to the target machines. Make
settings so that monitored machines can be accessed with their machine names.
1.10.3 Timing of Applying of Configuration Information
The system configuration information of SystemProvisioning can be applied manually or automatically. Using
the automatic applying function, System Monitor - Performance Monitoring Services communicates with
SystemProvisioning at regular intervals, and automatically updates any changes to the configuration
information. By default, communication with SystemProvisioning is performed every 10 minutes to
automatically apply the configuration information.
1.10.4 Rules for Applying Configuration Information
(1) Applying by the performance monitoring setting in the group or model of SigmaSystemCenter
When the performance data collection settings for groups or models of SigmaSystemCenter is enabled,
Applying the configuration information of SystemProvisioning in System Monitor - Performance
Monitoring Services adds groups, machines, performance indicators to be collected and thresholds settings
to be monitored or applies machine status to System Monitor - Performance Monitoring Services.
Applying the group/machine information
When the performance data collection settings for groups or models of SigmaSystemCenter is enabled,
applying the configuration information of SystemProvisioning in System Monitor - Performance
Monitoring Services adds groups and machines which belong to the groups.
Note
43
Configuration information comparison
(whether monitoring profiles is configured)
Content applied to System Monitor - Performance Monitoring Services
Notes SigmaSystemCenter
group/model
System Monitor - Performance
Monitoring Services group
Add/Delete a group
Add/Delete a
machine in
the group
Performance data
collecting operation
Configure Not exist Add groups Add machines
Started If performance monitoring is enabled on SigmaSystemCenter but no corresponding group exists in System Monitor - Performance Monitoring Services, a new group is added to System Monitor - Performance Monitoring Services and System Monitor - Performance Monitoring Services starts monitoring for the group.
Configure
Exist No changes No changes
Started If performance monitoring for a group is enabled on SigmaSystemCenter, System Monitor - Performance Monitoring Services starts monitoring for its corresponding group.
Not
Configure
Exist Not delete Not delete
Stopped If performance monitoring is disabled on SigmaSystemCenter, System Monitor - Performance Monitoring Services stops performance monitoring.
Applying the machine information
When the running state of the machine has been changed by executing Allocate Machine or Release
Machine on SigmaSystemCenter, apply the SystemProvisioning configuration information to System
Monitor - Performance Monitoring Services to add a machine or apply the machine state.
44
Configuration information comparison
(whether machines exist) Content applied to System Monitor -
Performance Monitoring Services
Notes Machines on SigmaSystemCenter
Health state
Machines on System Monitor -
Performance Monitoring
Services
Add/Delete a machine
Performance data
collecting operation
Active Normal Not monitored Add Started This machine exists
only in the
configuration
information of
SigmaSystemCenter.
Error Stopped
Active Normal Monitored No changes Started This machine exists in
the configuration
information of both
SigmaSystemCenter
and System Monitor -
Performance
Monitoring Services.
Error Stopped
Not active - Not monitored Delete Stopped This machine exists
only in the
configuration
information of System
Monitor - Performance
Monitoring Services.
Applying the performance information to be collected
System Monitor - Performance Monitoring Services adds performance indicators to be collected based on
the performance indicator settings in the monitoring profile specified in SigmaSystemCenter while
applying the configuration information of SystemProvisioning.
45
Configuration information comparison
(whether machines exist) Content applied to System Monitor - Performance Monitoring Services
Notes Is the performance indicator configured in the monitoring profile?
Is the performance indicator collected by
System Monitor - Performance
Monitoring Services?
Add/Delete the performance indicator to be collected
Yes No Add The performance
indicator is added on the
monitoring profile of
SigmaSystemCenter.
Yes Yes Change (such as changing a
collecting interval)
The performance
indicator is changed on
the monitoring profile of
SigmaSystemCenter.
No Yes Delete The performance
indicator is deleted from
the monitoring profile of
SigmaSystemCenter.
Applying the threshold information to be monitored
System Monitor - Performance Monitoring Services adds threshold information to be monitored based on
the threshold settings in the monitoring profile specified in SigmaSystemCenter while applying the
configuration information of SystemProvisioning.
46
Configuration information comparison
(whether machines exist)
Content applied to System Monitor - Performance
Monitoring Services
Notes Is the threshold information
configured in the monitoring
profile?
Is the threshold
monitored by System
Monitor - Performance
Monitoring Services?
Add/Delete the threshold
information to be monitored
Yes No Add The threshold monitoring
information is added on the
monitoring profile of
SigmaSystemCenter.
Yes Yes Change (frequency and
sensitivity of notification)
The threshold monitoring
information is changed on
the monitoring profile of
SigmaSystemCenter.
No Yes Delete The threshold monitoring
information is disabled on or
deleted from the monitoring
profile of
SigmaSystemCenter.
(2) Applying by the performance monitoring setting of System Monitor - Performance Monitoring Services
If the specified group/model exists in the SigmaSystemCenter configuration information, changes such as
machine additions and changes to the operational states of machines are applied. If the specified
group/model does not exist, an error will be noticed and configuration information of this group will not be
applied.
Configuration information comparison
(whether machines exist) Content applied to System Monitor -
Performance Monitoring Services Notes
SystemProvisioning group/model Health state
System Monitor - Performance Monitoring
Services group
Machine added or machine deleted
Performance data
collecting operation
Exists Normal Does not exist Machine added Started Servers that only exist in the configuration information for SigmaSystemCenter
Error Stopped
Exists Normal Exists No change Started Servers in the configuration information for both products Error Stopped
Does not exist - Exists Machine deleted Stopped Machines that only exist in the configuration information for System Monitor - Performance Monitoring Services
47
The operational states for machines with SigmaSystemCenter are as follows: Normal The state of machines that meet all of the following conditions for display in SigmaSystemCenter
Web Console: Power Status: "On" OS Status: "On" Executing Status: Excluding "Executing"
Error All other states
In applying configuration information of SystemProvisioning, a performance data collecting operation for the machine
which is in a normal state will be still pausing when the collecting status for the machine has already been
changed to paused status manually.
The information of machine status, hardware status and machine type applied from SystemProvisioning
configuration is shown in the Summary tab on the Machine Settings window.
1.10.5 Configuring Setting Items that cannot be Set in the Performance Monitoring Setting in
SigmaSystemCenter
You cannot configure the SigmaSystemCenter performance monitoring settings per host because those settings
are for the group or model properties only. So, you cannot use the SigmaSystemCenter Web console to change
the per-host monitoring settings if those settings need to be changed.
In this case, you need to configure the per-host monitoring settings through the System Monitor - Performance
Monitoring Services management console.
The following table shows whether you are allowed to change, delete or configure the settings individually.
Type Whether it is allowed to change, delete, or configure individually
Collecting data setting The collecting data settings automatically applied to System Monitor -
Performance Monitoring Services according to the SigmaSystemCenter
settings cannot be changed and deleted from System Monitor -
Performance Monitoring Services.
You can add new collecting data settings to a group on System Monitor -
Performance Monitoring Services. You can also configure the data
collection settings individually for monitored machines.
For how to add data collection settings, see Subsection 4.2.2, "How to
Specify Collecting Data Settings."
See "10.4 Notes on Linking to SystemProvisioning" for information on the
SystemProvisioning configuration information applying function. Note
48
Type Whether it is allowed to change, delete, or configure individually
Threshold monitor setting Only the threshold monitor settings which have been automatically
applied to System Monitor - Performance Monitoring Services according
to the VM Optimized Placement settings can be changed on System
Monitor - Performance Monitoring Services. Otherwise, you cannot
change or delete the threshold monitor settings which have been
automatically applied to System Monitor - Performance Monitoring
Services according to the SigmaSystemCenter settings on System
Monitor - Performance Monitoring Services.
You can add new threshold monitor settings on System Monitor -
Performance Monitoring Services. For how to add threshold monitoring
settings, see Subsection 6.1, "Configuring Threshold Monitoring."
Connection setting The connection settings for a group which have been automatically
applied to System Monitor - Performance Monitoring Services according
to the SigmaSystemCenter settings can be changed on System Monitor -
Performance Monitoring Services. You can also configure connection
settings individually for machines within the group. For how to set up
connection, see Subsection 4.1, "Specifications for Monitored Machines."
1.10.6 Notifying Performance Abnormalities
Load status abnormalities with monitored machines that are detected by System Monitor - Performance
Monitoring Services's threshold monitoring can be notified to SigmaSystemCenter. SigmaSystemCenter
receives these notifications and executes actions (such as adding machines) according to its policies.
Notifications can be made at over-threshold and at recovery. For notification, set notification category to notice.
As actions according to policy cannot be set for SystemProvisioning category groups, when category groups are
assigned to System Monitor - Performance Monitoring Services, they are disabled even when threshold
definitions are set for groups. To link the recovery process to threshold monitoring, assign the
SystemProvisioning groups or subgroups to groups of System Monitor - Performance Monitoring Services.
(1) Notification Categories
"Notification categories" are the different classifications for associating the System Monitor - Performance
Monitoring Services events that are sent when over-threshold conditions occur or recovery from
over-threshold conditions occur with the SigmaSystemCenter recovery processes that are triggered by
these events. There are two types of notification categories with the different type of performance indicator
that can be set. It is possible to set built-in notification categories and customized notification categories
both for built-in performance indicators. But only customized notification categories can set for customized
performance indicators.
49
1. Built-In Notification Categories
There are eight types of built-in notification categories for each built-in performance indicator. It is
possible to set the recovery action from the policy setting in SigmaSystemCenter.
The types of built-in notification categories are listed as follows:
① Upper Error Exceeded: The event is notified to the SigmaSystemCenter when the collected
performance data reaches or exceeds the upper-limit critical value.
② Upper Error Recovery: The event is notified to the SigmaSystemCenter when the collected
performance data drops to or below the upper-limit critical value.
③ Upper Warning Exceeded: The event is notified to the SigmaSystemCenter when the collected
performance data reaches or exceeds the upper-limit warning value.
④ Upper Warning Recovery: The event is notified to the SigmaSystemCenter when the collected
performance data drops to or below the upper-limit warning value.
⑤ Lower Warning Exceeded: The event is notified to the SigmaSystemCenter when the collected
performance data drops to or below the lower-limit warning value.
⑥ Lower Warning Recovery: The event is notified to the SigmaSystemCenter when the collected
performance data reaches or exceeds the lower-limit warning value.
⑦ Lower Error Exceeded: The event is notified to the SigmaSystemCenter when the collected
performance data drops to or below the lower-limit critical value.
⑧ Lower Error Recovery: The event is notified to the SigmaSystemCenter when the collected
performance data reaches or exceeds the lower-limit critical value.
The built-in notification categories are listed as follows:
Performance
Indicator Event ID Event
CPU Usage
(%)
0x20000100 CPU Usage (%) Lower Warning Recovery
0x20000101 CPU Usage (%) Lower Warning Exceeded
0x20000102 CPU Usage (%) Lower Error Recovery
0x20000103 CPU Usage (%) Lower Error Exceeded
0x20000104 CPU Usage (%) Upper Warning Recovery
0x20000105 CPU Usage (%) Upper Warning Exceeded
0x20000106 CPU Usage (%) Upper Error Recovery
0x20000107 CPU Usage (%) Upper Error Exceeded
CPU System
Usage (%)
0x20000200 CPU System Usage (%) Lower Warning Recovery
0x20000201 CPU System Usage (%) Lower Warning Exceeded
0x20000202 CPU System Usage (%) Lower Error Recovery
0x20000203 CPU System Usage (%) Lower Error Exceeded
0x20000204 CPU System Usage (%) Upper Warning Recovery
0x20000205 CPU System Usage (%) Upper Warning Exceeded
0x20000206 CPU System Usage (%) Upper Error Recovery
50
Performance
Indicator Event ID Event
0x20000207 CPU System Usage (%) Upper Error Exceeded
CPU User
Usage (%)
0x20000300 CPU User Usage (%) Lower Warning Recovery
0x20000301 CPU User Usage (%) Lower Warning Exceeded
0x20000302 CPU User Usage (%) Lower Error Recovery
0x20000303 CPU User Usage (%) Lower Error Exceeded
0x20000304 CPU User Usage (%) Upper Warning Recovery
0x20000305 CPU User Usage (%) Upper Warning Exceeded
0x20000306 CPU User Usage (%) Upper Error Recovery
0x20000307 CPU User Usage (%) Upper Error Exceeded
CPU Usage
(MHz)
0x20000400 CPU Usage (MHz) Lower Warning Recovery
0x20000401 CPU Usage (MHz) Lower Warning Exceeded
0x20000402 CPU Usage (MHz) Lower Error Recovery
0x20000403 CPU Usage (MHz) Lower Error Exceeded
0x20000404 CPU Usage (MHz) Upper Warning Recovery
0x20000405 CPU Usage (MHz) Upper Warning Exceeded
0x20000406 CPU Usage (MHz) Upper Error Recovery
0x20000407 CPU Usage (MHz) Upper Error Exceeded
Guest CPU
Usage (%)
0x20000B00 Guest CPU Usage (%) Lower Warning Recovery
0x20000B01 Guest CPU Usage (%) Lower Warning Exceeded
0x20000B02 Guest CPU Usage (%) Lower Error Recovery
0x20000B03 Guest CPU Usage (%) Lower Error Exceeded
0x20000B04 Guest CPU Usage (%) Upper Warning Recovery
0x20000B05 Guest CPU Usage (%) Upper Warning Exceeded
0x20000B06 Guest CPU Usage (%) Upper Error Recovery
0x20000B07 Guest CPU Usage (%) Upper Error Exceeded
Guest CPU
Usage (MHz)
0x20000C00 Guest CPU Usage (MHz) Lower Warning Recovery
0x20000C01 Guest CPU Usage (MHz) Lower Warning Exceeded
0x20000C02 Guest CPU Usage (MHz) Lower Error Recovery
0x20000C03 Guest CPU Usage (MHz) Lower Error Exceeded
0x20000C04 Guest CPU Usage (MHz) Upper Warning Recovery
0x20000C05 Guest CPU Usage (MHz) Upper Warning Exceeded
0x20000C06 Guest CPU Usage (MHz) Upper Error Recovery
0x20000C07 Guest CPU Usage (MHz) Upper Error Exceeded
Host CPU
Usage (%)
0x20000D00 Host CPU Usage (%) Lower Warning Recovery
0x20000D01 Host CPU Usage (%) Lower Warning Exceeded
0x20000D02 Host CPU Usage (%) Lower Error Recovery
51
Performance
Indicator Event ID Event
0x20000D03 Host CPU Usage (%) Lower Error Exceeded
0x20000D04 Host CPU Usage (%) Upper Warning Recovery
0x20000D05 Host CPU Usage (%) Upper Warning Exceeded
0x20000D06 Host CPU Usage (%) Upper Error Recovery
0x20000D07 Host CPU Usage (%) Upper Error Exceeded
Host CPU
Usage (MHz)
0x20000E00 Host CPU Usage (MHz) Lower Warning Recovery
0x20000E01 Host CPU Usage (MHz) Lower Warning Exceeded
0x20000E02 Host CPU Usage (MHz) Lower Error Recovery
0x20000E03 Host CPU Usage (MHz) Lower Error Exceeded
0x20000E04 Host CPU Usage (MHz) Upper Warning Recovery
0x20000E05 Host CPU Usage (MHz) Upper Warning Exceeded
0x20000E06 Host CPU Usage (MHz) Upper Error Recovery
0x20000E07 Host CPU Usage (MHz) Upper Error Exceeded
Disk Transfer
Rate
(Bytes/sec)
0x20001500 Disk Transfer Rate (Bytes/sec) Lower Warning Recovery
0x20001501 Disk Transfer Rate (Bytes/sec) Lower Warning Exceeded
0x20001502 Disk Transfer Rate (Bytes/sec) Lower Error Recovery
0x20001503 Disk Transfer Rate (Bytes/sec) Lower Error Exceeded
0x20001504 Disk Transfer Rate (Bytes/sec) Upper Warning Recovery
0x20001505 Disk Transfer Rate (Bytes/sec) Upper Warning Exceeded
0x20001506 Disk Transfer Rate (Bytes/sec) Upper Error Recovery
0x20001507 Disk Transfer Rate (Bytes/sec) Upper Error Exceeded
Disk IO
Count
(IO/sec)
0x20001600 Disk IO Count (IO/sec) Lower Warning Recovery
0x20001601 Disk IO Count (IO/sec) Lower Warning Exceeded
0x20001602 Disk IO Count (IO/sec) Lower Error Recovery
0x20001603 Disk IO Count (IO/sec) Lower Error Exceeded
0x20001604 Disk IO Count (IO/sec) Upper Warning Recovery
0x20001605 Disk IO Count (IO/sec) Upper Warning Exceeded
0x20001606 Disk IO Count (IO/sec) Upper Error Recovery
0x20001607 Disk IO Count (IO/sec) Upper Error Exceeded
Disk Read
Transfer Rate
(Bytes/sec)
0x20001700 Disk Read Transfer Rate (Bytes/sec) Lower Warning Recovery
0x20001701 Disk Read Transfer Rate (Bytes/sec) Lower Warning Exceeded
0x20001702 Disk Read Transfer Rate (Bytes/sec) Lower Error Recovery
0x20001703 Disk Read Transfer Rate (Bytes/sec) Lower Error Exceeded
0x20001704 Disk Read Transfer Rate (Bytes/sec) Upper Warning Recovery
0x20001705 Disk Read Transfer Rate (Bytes/sec) Upper Warning Exceeded
0x20001706 Disk Read Transfer Rate (Bytes/sec) Upper Error Recovery
52
Performance
Indicator Event ID Event
0x20001707 Disk Read Transfer Rate (Bytes/sec) Upper Error Exceeded
Disk Read
Count
(IO/sec)
0x20001800 Disk Read Count (IO/sec) Lower Warning Recovery
0x20001801 Disk Read Count (IO/sec) Lower Warning Exceeded
0x20001802 Disk Read Count (IO/sec) Lower Error Recovery
0x20001803 Disk Read Count (IO/sec) Lower Error Exceeded
0x20001804 Disk Read Count (IO/sec) Upper Warning Recovery
0x20001805 Disk Read Count (IO/sec) Upper Warning Exceeded
0x20001806 Disk Read Count (IO/sec) Upper Error Recovery
0x20001807 Disk Read Count (IO/sec) Upper Error Exceeded
Disk Write
Transfer Rate
(Bytes/sec)
0x20001900 Disk Write Transfer Rate (Bytes/sec) Lower Warning Recovery
0x20001901 Disk Write Transfer Rate (Bytes/sec) Lower Warning Exceeded
0x20001902 Disk Write Transfer Rate (Bytes/sec) Lower Error Recovery
0x20001903 Disk Write Transfer Rate (Bytes/sec) Lower Error Exceeded
0x20001904 Disk Write Transfer Rate (Bytes/sec) Upper Warning Recovery
0x20001905 Disk Write Transfer Rate (Bytes/sec) Upper Warning Exceeded
0x20001906 Disk Write Transfer Rate (Bytes/sec) Upper Error Recovery
0x20001907 Disk Write Transfer Rate (Bytes/sec) Upper Error Exceeded
Disk Write
Count
(IO/sec)
0x20001A00 Disk Write Count (IO/sec) Lower Warning Recovery
0x20001A01 Disk Write Count (IO/sec) Lower Warning Exceeded
0x20001A02 Disk Write Count (IO/sec) Lower Error Recovery
0x20001A03 Disk Write Count (IO/sec) Lower Error Exceeded
0x20001A04 Disk Write Count (IO/sec) Upper Warning Recovery
0x20001A05 Disk Write Count (IO/sec) Upper Warning Exceeded
0x20001A06 Disk Write Count (IO/sec) Upper Error Recovery
0x20001A07 Disk Write Count (IO/sec) Upper Error Exceeded
Disk Space
(MB)
0x20001B00 Disk Space (MB) Lower Warning Recovery
0x20001B01 Disk Space (MB) Lower Warning Exceeded
0x20001B02 Disk Space (MB) Lower Error Recovery
0x20001B03 Disk Space (MB) Lower Error Exceeded
0x20001B04 Disk Space (MB) Upper Warning Recovery
0x20001B05 Disk Space (MB) Upper Warning Exceeded
0x20001B06 Disk Space (MB) Upper Error Recovery
0x20001B07 Disk Space (MB) Upper Error Exceeded
Disk Space
Ratio (%)
0x20001C00 Disk Space Ratio (%) Lower Warning Recovery
0x20001C01 Disk Space Ratio (%) Lower Warning Exceeded
0x20001C02 Disk Space Ratio (%) Lower Error Recovery
53
Performance
Indicator Event ID Event
0x20001C03 Disk Space Ratio (%) Lower Error Exceeded
0x20001C04 Disk Space Ratio (%) Upper Warning Recovery
0x20001C05 Disk Space Ratio (%) Upper Warning Exceeded
0x20001C06 Disk Space Ratio (%) Upper Error Recovery
0x20001C07 Disk Space Ratio (%) Upper Error Exceeded
Guest Disk
Transfer Rate
(Bytes/sec)
0x20001F00 Guest Disk Transfer Rate (Bytes/sec) Lower Warning Recovery
0x20001F01 Guest Disk Transfer Rate (Bytes/sec) Lower Warning Exceeded
0x20001F02 Guest Disk Transfer Rate (Bytes/sec) Lower Error Recovery
0x20001F03 Guest Disk Transfer Rate (Bytes/sec) Lower Error Exceeded
0x20001F04 Guest Disk Transfer Rate (Bytes/sec) Upper Warning Recovery
0x20001F05 Guest Disk Transfer Rate (Bytes/sec) Upper Warning Exceeded
0x20001F06 Guest Disk Transfer Rate (Bytes/sec) Upper Error Recovery
0x20001F07 Guest Disk Transfer Rate (Bytes/sec) Upper Error Exceeded
Guest Disk
IO Count
(IO/sec)
0x20002000 Guest Disk IO Count (IO/sec) Lower Warning Recovery
0x20002001 Guest Disk IO Count (IO/sec) Lower Warning Exceeded
0x20002002 Guest Disk IO Count (IO/sec) Lower Error Recovery
0x20002003 Guest Disk IO Count (IO/sec) Lower Error Exceeded
0x20002004 Guest Disk IO Count (IO/sec) Upper Warning Recovery
0x20002005 Guest Disk IO Count (IO/sec) Upper Warning Exceeded
0x20002006 Guest Disk IO Count (IO/sec) Upper Error Recovery
0x20002007 Guest Disk IO Count (IO/sec) Upper Error Exceeded
Guest Disk
Read Transfer
Rate
(Bytes/sec)
0x20002100 Guest Disk Read Transfer Rate (Bytes/sec) Lower Warning
Recovery
0x20002101 Guest Disk Read Transfer Rate (Bytes/sec) Lower Warning
Exceeded
0x20002102 Guest Disk Read Transfer Rate (Bytes/sec) Lower Error
Recovery
0x20002103 Guest Disk Read Transfer Rate (Bytes/sec) Lower Error
Exceeded
0x20002104 Guest Disk Read Transfer Rate (Bytes/sec) Upper Warning
Recovery
0x20002105 Guest Disk Read Transfer Rate (Bytes/sec) Upper Warning
Exceeded
0x20002106 Guest Disk Read Transfer Rate (Bytes/sec) Upper Error
Recovery
0x20002107 Guest Disk Read Transfer Rate (Bytes/sec) Upper Error
54
Performance
Indicator Event ID Event
Exceeded
Guest Disk
Read Count
(IO/sec)
0x20002200 Guest Disk Read Count (IO/sec) Lower Warning Recovery
0x20002201 Guest Disk Read Count (IO/sec) Lower Warning Exceeded
0x20002202 Guest Disk Read Count (IO/sec) Lower Error Recovery
0x20002203 Guest Disk Read Count (IO/sec) Lower Error Exceeded
0x20002204 Guest Disk Read Count (IO/sec) Upper Warning Recovery
0x20002205 Guest Disk Read Count (IO/sec) Upper Warning Exceeded
0x20002206 Guest Disk Read Count (IO/sec) Upper Error Recovery
0x20002207 Guest Disk Read Count (IO/sec) Upper Error Exceeded
Guest Disk
Write
Transfer Rate
(Bytes/sec)
0x20002300 Guest Disk Write Transfer Rate (Bytes/sec) Lower Warning
Recovery
0x20002301 Guest Disk Write Transfer Rate (Bytes/sec) Lower Warning
Exceeded
0x20002302 Guest Disk Write Transfer Rate (Bytes/sec) Lower Error
Recovery
0x20002303 Guest Disk Write Transfer Rate (Bytes/sec) Lower Error
Exceeded
0x20002304 Guest Disk Write Transfer Rate (Bytes/sec) Upper Warning
Recovery
0x20002305 Guest Disk Write Transfer Rate (Bytes/sec) Upper Warning
Exceeded
0x20002306 Guest Disk Write Transfer Rate (Bytes/sec) Upper Error
Recovery
0x20002307 Guest Disk Write Transfer Rate (Bytes/sec) Upper Error
Exceeded
Guest Disk
Write Count
(IO/sec)
0x20002400 Guest Disk Write Count (IO/sec) Lower Warning Recovery
0x20002401 Guest Disk Write Count (IO/sec) Lower Warning Exceeded
0x20002402 Guest Disk Write Count (IO/sec) Lower Error Recovery
0x20002403 Guest Disk Write Count (IO/sec) Lower Error Exceeded
0x20002404 Guest Disk Write Count (IO/sec) Upper Warning Recovery
0x20002405 Guest Disk Write Count (IO/sec) Upper Warning Exceeded
0x20002406 Guest Disk Write Count (IO/sec) Upper Error Recovery
0x20002407 Guest Disk Write Count (IO/sec) Upper Error Exceeded
Guest Disk
Usage (MB)
0x20002500 Guest Disk Usage (MB) Lower Warning Recovery
0x20002501 Guest Disk Usage (MB) Lower Warning Exceeded
0x20002502 Guest Disk Usage (MB) Lower Error Recovery
55
Performance
Indicator Event ID Event
0x20002503 Guest Disk Usage (MB) Lower Error Exceeded
0x20002504 Guest Disk Usage (MB) Upper Warning Recovery
0x20002505 Guest Disk Usage (MB) Upper Warning Exceeded
0x20002506 Guest Disk Usage (MB) Upper Error Recovery
0x20002507 Guest Disk Usage (MB) Upper Error Exceeded
Guest Disk
Usage (%)
0x20002600 Guest Disk Usage (%) Lower Warning Recovery
0x20002601 Guest Disk Usage (%) Lower Warning Exceeded
0x20002602 Guest Disk Usage (%) Lower Error Recovery
0x20002603 Guest Disk Usage (%) Lower Error Exceeded
0x20002604 Guest Disk Usage (%) Upper Warning Recovery
0x20002605 Guest Disk Usage (%) Upper Warning Exceeded
0x20002606 Guest Disk Usage (%) Upper Error Recovery
0x20002607 Guest Disk Usage (%) Upper Error Exceeded
Network
Packet
Transfer Rate
(Bytes/sec)
0x20002900 Network Packet Transfer Rate (Bytes/sec) Lower Warning
Recovery
0x20002901 Network Packet Transfer Rate (Bytes/sec) Lower Warning
Exceeded
0x20002902 Network Packet Transfer Rate (Bytes/sec) Lower Error
Recovery
0x20002903 Network Packet Transfer Rate (Bytes/sec) Lower Error
Exceeded
0x20002904 Network Packet Transfer Rate (Bytes/sec) Upper Warning
Recovery
0x20002905 Network Packet Transfer Rate (Bytes/sec) Upper Warning
Exceeded
0x20002906 Network Packet Transfer Rate (Bytes/sec) Upper Error
Recovery
0x20002907 Network Packet Transfer Rate (Bytes/sec) Upper Error
Exceeded
Network
Packet
Reception
Rate
(Bytes/sec)
0x20002A00 Network Packet Reception Rate (Bytes/sec) Lower Warning
Recovery
0x20002A01 Network Packet Reception Rate (Bytes/sec) Lower Warning
Exceeded
0x20002A02 Network Packet Reception Rate (Bytes/sec) Lower Error
Recovery
0x20002A03 Network Packet Reception Rate (Bytes/sec) Lower Error
56
Performance
Indicator Event ID Event
Exceeded
0x20002A04 Network Packet Reception Rate (Bytes/sec) Upper Warning
Recovery
0x20002A05 Network Packet Reception Rate (Bytes/sec) Upper Warning
Exceeded
0x20002A06 Network Packet Reception Rate (Bytes/sec) Upper Error
Recovery
0x20002A07 Network Packet Reception Rate (Bytes/sec) Upper Error
Exceeded
Network
Packet
Transmission
Rate
(Bytes/sec)
0x20002B00 Network Packet Transmission Rate (Bytes/sec) Lower Warning
Recovery
0x20002B01 Network Packet Transmission Rate (Bytes/sec) Lower Warning
Exceeded
0x20002B02 Network Packet Transmission Rate (Bytes/sec) Lower Error
Recovery
0x20002B03 Network Packet Transmission Rate (Bytes/sec) Lower Error
Exceeded
0x20002B04 Network Packet Transmission Rate (Bytes/sec) Upper Warning
Recovery
0x20002B05 Network Packet Transmission Rate (Bytes/sec) Upper Warning
Exceeded
0x20002B06 Network Packet Transmission Rate (Bytes/sec) Upper Error
Recovery
0x20002B07 Network Packet Transmission Rate (Bytes/sec) Upper Error
Exceeded
Guest
Network
Transfer Rate
(Bytes/sec)
0x20003300 Guest Network Transfer Rate (Bytes/sec) Lower Warning
Recovery
0x20003301 Guest Network Transfer Rate (Bytes/sec) Lower Warning
Exceeded
0x20003302 Guest Network Transfer Rate (Bytes/sec) Lower Error Recovery
0x20003303 Guest Network Transfer Rate (Bytes/sec) Lower Error Exceeded
0x20003304 Guest Network Transfer Rate (Bytes/sec) Upper Warning
Recovery
0x20003305 Guest Network Transfer Rate (Bytes/sec) Upper Warning
Exceeded
0x20003306 Guest Network Transfer Rate (Bytes/sec) Upper Error Recovery
57
Performance
Indicator Event ID Event
0x20003307 Guest Network Transfer Rate (Bytes/sec) Upper Error
Exceeded
Physical
Memory
Space (MB)
0x20003D00 Physical Memory Space (MB) Lower Warning Recovery
0x20003D01 Physical Memory Space (MB) Lower Warning Exceeded
0x20003D02 Physical Memory Space (MB) Lower Error Recovery
0x20003D03 Physical Memory Space (MB) Lower Error Exceeded
0x20003D04 Physical Memory Space (MB) Upper Warning Recovery
0x20003D05 Physical Memory Space (MB) Upper Warning Exceeded
0x20003D06 Physical Memory Space (MB) Upper Error Recovery
0x20003D07 Physical Memory Space (MB) Upper Error Exceeded
Physical
Memory
Space Ratio
(%)
0x20003E00 Physical Memory Space Ratio (%) Lower Warning Recovery
0x20003E01 Physical Memory Space Ratio (%) Lower Warning Exceeded
0x20003E02 Physical Memory Space Ratio (%) Lower Error Recovery
0x20003E03 Physical Memory Space Ratio (%) Lower Error Exceeded
0x20003E04 Physical Memory Space Ratio (%) Upper Warning Recovery
0x20003E05 Physical Memory Space Ratio (%) Upper Warning Exceeded
0x20003E06 Physical Memory Space Ratio (%) Upper Error Recovery
0x20003E07 Physical Memory Space Ratio (%) Upper Error Exceeded
Guest
Memory
Usage (%)
0x20004700 Guest Memory Usage (%) Lower Warning Recovery
0x20004701 Guest Memory Usage (%) Lower Warning Exceeded
0x20004702 Guest Memory Usage (%) Lower Error Recovery
0x20004703 Guest Memory Usage (%) Lower Error Exceeded
0x20004704 Guest Memory Usage (%) Upper Warning Recovery
0x20004705 Guest Memory Usage (%) Upper Warning Exceeded
0x20004706 Guest Memory Usage (%) Upper Error Recovery
0x20004707 Guest Memory Usage (%) Upper Error Exceeded
Guest
Memory
Usage (MB)
0x20004800 Guest Memory Usage (MB) Lower Warning Recovery
0x20004801 Guest Memory Usage (MB) Lower Warning Exceeded
0x20004802 Guest Memory Usage (MB) Lower Error Recovery
0x20004803 Guest Memory Usage (MB) Lower Error Exceeded
0x20004804 Guest Memory Usage (MB) Upper Warning Recovery
0x20004805 Guest Memory Usage (MB) Upper Warning Exceeded
0x20004806 Guest Memory Usage (MB) Upper Error Recovery
0x20004807 Guest Memory Usage (MB) Upper Error Exceeded
Host Memory
Usage (%)
0x20004900 Host Memory Usage (%) Lower Warning Recovery
0x20004901 Host Memory Usage (%) Lower Warning Exceeded
58
Performance
Indicator Event ID Event
0x20004902 Host Memory Usage (%) Lower Error Recovery
0x20004903 Host Memory Usage (%) Lower Error Exceeded
0x20004904 Host Memory Usage (%) Upper Warning Recovery
0x20004905 Host Memory Usage (%) Upper Warning Exceeded
0x20004906 Host Memory Usage (%) Upper Error Recovery
0x20004907 Host Memory Usage (%) Upper Error Exceeded
Host Memory
Usage (MB)
0x20004A00 Host Memory Usage (MB) Lower Warning Recovery
0x20004A01 Host Memory Usage (MB) Lower Warning Exceeded
0x20004A02 Host Memory Usage (MB) Lower Error Recovery
0x20004A03 Host Memory Usage (MB) Lower Error Exceeded
0x20004A04 Host Memory Usage (MB) Upper Warning Recovery
0x20004A05 Host Memory Usage (MB) Upper Warning Exceeded
0x20004A06 Host Memory Usage (MB) Upper Error Recovery
0x20004A07 Host Memory Usage (MB) Upper Error Exceeded
Current
Power (W)
0x20006500 Current Power (W) Lower Warning Recovery
0x20006501 Current Power (W) Lower Warning Exceeded
0x20006502 Current Power (W) Lower Error Recovery
0x20006503 Current Power (W) Lower Error Exceeded
0x20006504 Current Power (W) Upper Warning Recovery
0x20006505 Current Power (W) Upper Warning Exceeded
0x20006506 Current Power (W) Upper Error Recovery
0x20006507 Current Power (W) Upper Error Exceeded
2. Customized Notification Categories
There are two customized notification categories, depending on the nature of the
SigmaSystemCenter recovery process: a notification category for machines, and a notification
category for groups.
The customized notification categories are listed as follows:
Event Type Event ID Event
CustomizedMachineNotification 0x10000001 Customized machine notification1
0x10000002 Customized machine notification2
0x10000003 Customized machine notification3
0x10000004 Customized machine notification4
0x10000005 Customized machine notification5
0x10000006 Customized machine notification6
0x10000007 Customized machine notification7
59
Event Type Event ID Event
0x10000008 Customized machine notification8
0x10000009 Customized machine notification9
0x1000000A Customized machine notification10
CustomizedGroupNotification 0x11000001 Customized group notification1
0x11000002 Customized group notification2
0x11000003 Customized group notification3
0x11000004 Customized group notification4
0x11000005 Customized group notification5
With the "machine" notification category, recovery processes for individual machines can be set up,
such as "shut down", "reboot", "replace" and so on. With the "group" notification category, recovery
processes for groups can be set up, such as "add machine" or "delete machine" and so on.
When the threshold method is used with target nodes, the threshold method for which the "group"
customized notification category has been set cannot be associated with machine nodes. Similarly,
the threshold method for which the "machine" customized notification category has been set
cannot be associated with group nodes.
Note also that customized notification categories can also be used from other components, and so
uses may overlap if SystemProvisioning receives notifications from more than one component.
As actions according to policy cannot be set for SigmaSystemCenter category groups, when
category group is allocated to System Monitor - Performance Monitoring Services group, they are
invalid even if threshold definitions are set for groups. To set the recovery process linked to
threshold monitoring, assign SystemProvisioning groups or subgroups as groups of System
Monitor - Performance Monitoring Services.
(2) Applying Threshold Setting Information Linked to Changes in System Configurations
With System Monitor - Performance Monitoring Services, threshold monitoring settings can be configured
for three types of monitoring units: "machine", " group" and "all machines in group". In situations such as
when SigmaSystemCenter adds a machine to a group as a result of a performance abnormality notification,
System Monitor - Performance Monitoring Services must also treat the added machine as being part of the
group, and must monitor the load status of the group using the new group configuration. To use the
threshold monitoring function for this kind of purpose, set the monitoring unit for threshold monitoring
setting as either "group" or "all machines in group".
The following table shows the status of the threshold monitoring setting when System Monitor -
Performance Monitoring Services configuration information is changed.
60
Threshold monitoring
unit before the configuration
change
Change made to the
configuration
State of the threshold monitoring settings after the configuration
change Notes
Machine Machine added
Threshold monitoring settings are not set for the added machine.
Machine removed
Threshold monitoring settings are removed.
Group
Machine added, removed or moved
-
(The threshold monitoring settings for the group are unchanged.)
Threshold monitoring is performed for the machines that belong to the group.
Group removed
The threshold monitoring settings for the group are removed.
All machines in group
Machine added
Threshold monitoring settings are set for the added machine.
Machine removed
Threshold monitoring settings are removed for the removed machine.
Machine moved
The threshold monitoring settings for the original group will be discarded, and the settings for the new group are set up.
Group removed
Threshold monitoring settings are removed for all the machines in the group.
(3) Operational Notes
Warnings and abnormalities may occur at the same time if the values set for abnormality and warning
thresholds are very close, or if the load status changes suddenly. With SigmaSystemCenter, configurations
can change automatically due to such processes as fault recovery, and during the configuration change, the
performance status may change to an unexpected high load. A high load may also be detected again before
the machine addition is completed. Conduct thorough operational planning before linking to
SystemProvisioning configuration change processing.
See "10.4 Notes on Linking to SystemProvisioning" for more information on
SystemProvisioning configuration changes triggered by threshold monitoring. Note
61
1.10.7 VM Optimization Linkage Function
SigmaSystemCenter executes a load balancing policy, which balances the load imposed by the grouped VM
servers. It also applies a power saving policy, which powers off any unnecessary VM servers, by applying the
VM optimization configuration to a VM server group and receiving a performance failure report from System
Monitor - Performance Monitoring Services.
When the configuration information for a VM server model to which the VM optimization configuration has
been applied in SystemProvisioning is reflected on System Monitor - Performance Monitoring Services,
threshold settings that are tailored to the VM optimization configuration set in SystemProvisioning are
automatically created as the configuration information is reflected. The threshold monitor settings are
registered for all the servers in the group for which reflection was performed.
Management server
Collecting performance data
System Monitor – Performance Monitoring Services
SystemProvisioning
Notifying performance abnormalities
Applying system configuration
VM Optimization policy
Executing recovery action
VM server group
62
The threshold settings to be registered automatically are as listed below.
<Threshold Method for VM Server high load>
Item Setting Remark
Threshold method name “[VMServer] <SystemProvisioning path> CPU
Usage (%) High Load threshold definition”
Unchangeable
Performance indicator CPU Usage (%) Unchangeable
Statistics Average value Unchangeable
Upper limit critical value High load bound value for the VM optimization
configuration specified by the model setting in
SigmaSystemCenter.
The notification category during excess is "VM
Server high load".
Changeable only from
SigmaSystemCenter
Upper limit warning value Not set. Unchangeable
Lower limit warning value Not set. Unchangeable
Lower limit critical value Not set. Unchangeable
Notification settings Default value
(A notification is issued if the threshold is
exceeded ten times in 10 checks. Another
notification is issued every 10 checks.)
Changeable
<Threshold Method for VM Server low load>
Item Setting Remark
Threshold method name “[VMServer] <SystemProvisioning path> CPU
Usage (%) Low Load threshold definition”
Unchangeable
Performance indicator CPU Usage (%) Unchangeable
Statistics Average value Unchangeable
Upper limit critical value Not set. Unchangeable
Upper limit warning value Not set. Unchangeable
Lower limit warning value Not set. Unchangeable
Lower limit critical value Low load bound value for the VM optimization
configuration specified by the model setting in
SigmaSystemCenter.
The notification category during excess is "VM
Server low load".
Changeable only from
SigmaSystemCenter
Notification settings Default value
(A notification is issued if the threshold is
exceeded 30 times in 30 checks. Another
notification is issued every 30 checks.)
Changeable
63
Notification categories "VM Server high load" and "VM Server low load",
which are used by the threshold method obtained from SigmaSystemCenter
and then registered, cannot be used when the threshold method is registered
manually.
Note
The SystemProvisioning path to be set in the server group of System Monitor
- Performance Monitoring Services must be a full path to the model to use the
VM Optimized Placement feature to register the target virtual machine
server.
Note
64
1.11 Command-Line Interface
The performance data collected by System Monitor - Performance Monitoring Services can be displayed on the
command prompt by using the command-line tool (ssc-perf.exe). This makes it possible to track the performance
trends without starting the management console.
1.11.1 System Configuration
This tool is installed on the management server. This tool is executable on the management server.
An administrator privilege is needed to run this tool.
Management server
Collecting performance data
Saving performance data
System Monitor - Performance Monitoring Services Performance Monitoring Service
Exporting Performance
data
Monitored machines
Command-Line Tool (ssc-perf)
Querying performance
data
If the User Account Control, UAC, is valid, you need to execute with the
Administrator mode. For example, right-click Command Prompt and click
Run as administrator to launch the ssc-perf commands.
Note
65
1.11.2 Data Format and Target
The data format is same as the format shown in "1.8.5 Output Graph Data to File". The data is queried by the
specified parameters (node, performance indicator, statistical method, period and so on). Node, performance
indicator and statistical method are required parameters. You can specify multiple values for only one of the
three required parameters.
For the detail of the usage of the tool, see "8.1 ssc-perf Command".
66
Chapter 2 Configuration Settings
2.1 Starting the Performance Monitoring Service
2.1.1 About the Performance Monitoring Service
In order to use System Monitor - Performance Monitoring Services functions, the Performance Monitoring
Service must have started on the management server. The Performance Monitoring Service resides in the
background as a Windows service. The Performance Monitoring Service startup type is "Automatic" by default,
so it runs automatically when the OS starts.
2.1.2 How to Start the Performance Monitoring Service Manually
The Performance Monitoring Service can be started and stopped manually. The procedure is as follows.
1. On the Start menu, point to the Control Panel and then double click Administrative Tools.
The Administrative Tools window appears.
2. Double-click "System Monitor Performance Monitoring Service" from the services list.
The System Monitor Performance Monitoring Service Properties window appears.
3. On the General tab, click Start. The Performance Monitoring Service will start.
67
2.2 Registering the Management Server
To register the management servers in the management console, complete the following steps:
1. Start the management console.
2. Select Add Management Server from the Tools menu. See “3.1.2 Starting the Management Console” for more
information about starting the management console. The Management Server window appears. Enter the required items,
and then click OK.
3. Enter the hostname or IP address of the connection destination management server from the keyboard for the Host name.
The User name and Password specified at management console startup are displayed for User name and Password.
Change these when using a connection account that differs for each management server. The default value for Port
Number (26200) is already set, so no input is required. See “2.4.1 Port Number” for more information on port numbers.
4. Click OK to add the management server to the management console. The management server name is displayed in the
navigation tree window. Connection to the management server is also performed.
You can delete registered management servers from the navigation tree window. The connection from the management console
to the management server is lost when the management server registration is deleted. However, Performance Monitoring
Service operations are unaffected at disconnection. The Performance Monitoring Service continues to monitor according to its
settings. For information on how to delete registered management servers, see “3.1.4 Specifying from the Navigation Tree
Window”.
The '\' character cannot be used in the Host name text box. Note
68
2.3 Connection Settings to Monitored Machines
The connection settings between the management server (Performance Monitoring Service) and monitored
machines are explained below.
2.3.1 Setting the Connection Account to the Monitored Machine
Set the account name and password so that the Performance Monitoring Service can access the monitored
machine. To register, complete the following steps:
1. Start the management console.
2. Connect to the management server. See “3.1.2 Starting the Management Console” for information about
starting the management console.
3. Right-click the management server name from the navigation tree window in the main window of the
management console and click Set Configuration.
A Configuration window with the name of the server appears.
4. In the Connection tab, specify a new account and password, and then click OK.
5. If the OS on the monitored machine is Linux is the monitored machine, select the correct connecting
options in the Protocol and Authentication Protocol boxes. The Protocol and Authentication Protocol
settings are ignored if the monitored machine is Citrix XenServer or VMware ESX/ESXi.
6. If Public key authentication is selected, then specify the path for the private key file in Key file path. The
private key is stored on the management server.
See “1.7 Connecting to Monitored Machines” for applicable rules for connection settings to Managed
Machines, general notes for settings and monitored machine side settings.
69
2.4 Connection Settings to the Management Console
The connection settings between the management server (Performance Monitoring Service) and management
console are explained below.
2.4.1 Port Number
System Monitor - Performance Monitoring Services uses port number 26200 for communications from the
management console to the Performance Monitoring Service, and port number 26202 for communicating from the
Performance Monitoring Service to the management console.
Performance Monitoring Service protocol Management console
26200 TCP assigned automatically
assigned automatically TCP 26202
If this port number is already being used by another product, this port number must be changed.
The procedure for changing the port number is as follows:
Changing the port number for the performance monitoring service
1. Start the management console, and then connect to the management server. See "3.1.2 Starting the Management Console"
for information on how to start the management console.
2. Right-click the management server name in the navigation tree window in the main window of the management console
and click Set configuration.
3. Click the Port tab.
4. In the New port number box, specify a new port number, and click OK.
70
The connection between the Performance Monitoring Service and the management console must be
restarted. The Performance Monitoring Service runs with the new port number when the Service is
restarted
Reconnect to the Performance Monitoring Service from the management console using the following procedure:
1. Right-click the management server name.
A popup menu appears.
2. Click Connection.
3. Click Set.
The Management Server window appears.
4. Specify the changed port number in the Port Number field and click OK to reconnect.
Changing the port number for the management console
1. Start the management console.
2. Click Set Management Console from the Tools menu.
The Management Console window appears.
3. In the Port number box, specify a new port number, and then click OK.
Note: When you click OK, the management console reconnects using the specified port number.
71
2.4.2 Firewall
If the management console is to be run on a machine other than the management server, the management
console and the Performance Monitoring Service will connect using the network.
If a firewall has been set up on the management server, then unblock the Performance Monitor Service.
Unblock settings can be made when System Monitor - Performance Monitoring Services is installed. If
unblock settings were not made when this product was installed, use the following procedure to do so: 1. On the Start menu, point to Control Panel, and then click Windows Firewall.
The Windows Firewall window appears. 2. Click the Exceptions tab. 3. Click Add Program.
The Add a Program window appears. 4. Add settings for the following program:
<installation directory>\bin\rm_pfmservice.exe
If a firewall has been set up on the management console machine, then unblock the management console
by using the following procedure: 1. On the Start menu, point to the Control Panel, and then click Windows Firewall.
The Windows Firewall window appears. 2. Click the Exceptions tab 3. Click Add Program.
The Add a Program window appears. 4. Add settings for the following program:
<installation directory>\bin\RM_PFMCONSOLE.exe
If Windows is blocking the program, the following window may appear when a user logs on. Unblock settings
can also be made using this window.
72
2.5 The Execution Account for the Performance Monitoring Service
2.5.1 Setting up the Performance Monitoring Service Execution Account
The Performance Monitoring Service runs as the local system account. This is the default setting. You can
change the execution account for the Performance Monitoring Service from the local system account and
configure it to use a specific account for the Performance Monitoring Service. If the OS for the monitored
machine is Windows 2000, the execution account must be changed from the local system account. See “2.5.2
How to Change the Performance Monitoring Service Execution Account” for changing the execution account.
If all of the following conditions are satisfied, the Performance Monitoring Service accesses the monitored
machine using the execution account:
♦ The operating system for the monitored machine is Windows.
♦ The execution account is changed from the local system account.
♦ The connection account for the monitored machines is not specified (account field is blank).
The execution account must satisfy the following conditions.
♦ The execution account for the Performance Monitoring Service must possess management server OS
administrator rights. There are also rights required for settings in order to start the Performance
Monitoring Service. See “2.5.3 Rights for the Performance Monitoring Service Execution Account” for rights
required and settings for the Performance Monitoring Service execution account.
♦ If the OS for the management server is Windows Server 2003, use an account that does not have a blank
password. If an account with a blank password is specified as the execution account for the Performance
Monitoring Service, the Performance Monitoring Service may not be able to start, depending on the security
settings for OS
73
2.5.2 How to Change the Performance Monitoring Service Execution Account
The procedure for changing the execution account is as follows:
1. Start Services from the Administrative Tools control panel.
2. Double-click "System Monitor Performance Monitoring Service" from the list of services.
3. In the Log On tab, set up the account and password.
In the example above, the "administrator" account for the management server is specified as the execution
account.
4. The new execution account will take affect when the Performance Monitoring Service is restarted.
74
2.5.3 Rights for the Performance Monitoring Service Execution Account
When using an account besides the local system account as the execution account, the Performance Monitoring
Service cannot start if the specified account does not have the following rights:
• Log on as a service
• Act as part of the OS (only if the OS is Windows 2000)
If the execution account is changed, then this right must been assigned. To verify that this right has been
assigned, complete the following steps:
1. On the Start menu, point to the Control Panel, and then click Administrative Tools.
The Local Security Settings window appears.
2. Double-click Local Security Policy.
The Local Security Settings window appears.
3. Double-click Local Policies.
4. Double-click User Rights Assignment.
75
5. Double-click Log on as a service.
The Log on as a service Properties window appears.
6. Verify that either the execution account for the Performance Monitoring Service or the group that the account belongs to
is displayed. If it is not displayed, add it by clicking the Add User or Group button.
7. If the OS is Windows 2000, perform the same check for Act as part of the operating system.
76
2.6 SystemProvisioning Connection Settings
SystemProvisioning information must be set in order to use the SystemProvisioning configuration applying
function or the performance abnormality notification function. You must configure these settings.
The procedure for configuring these settings is as follows:
1. Start the management console and connect to the management server. See "3.1.2 Starting the Management Console" for
information on how to start the management console.
2. Right-click the management server name from the navigation tree window in the main window of the management
console. A popup menu appears.
3. Click Set configuration.
The Configuration window appears.
4. Specify the Server name on which SystemProvisioning is installed from the SystemProvisioning tab. The default value is
a server name of the management server.
5. To automatically apply SystemProvisioning configuration information, check the Apply SystemProvisioning
Configuration automatically box, and set a polling interval between one and 1000 minutes. The default status for this
check box is checked. The default polling interval is 10 minutes.
6. To set up a polling start time, check the Specify polling start time box, and then set the polling start time.
7. Click OK.
77
If the Performance Monitoring Service and SystemProvisioning are not on the same management server,
following settings are necessary:
• Windows Firewall setting
If the Windows firewall has been enabled, make exception settings by using the following procedure:
1. On the Start menu, point to the Control Panel, and then click Windows Firewall.
The Windows Firewall window appears.
2. Click the Exceptions tab.
3. Click Add Port.
4. Add the settings for the following port. Specify any name for the Name field. Function added Protocol Port number SystemProvisioning Universal Connector TCP 26102
• Service execution account setting
The OS administrator account of the SystemProvisioning management server must be set as the execution account of System
Monitor Performance Monitoring Service. See "2.5 The Execution Account for the Performance Monitoring Service" for
information on how to set the service execution account.
When SystemProvisioning is clustered with EXPRESSCLUSTER, the virtual computer name cannot be used as
SystemProvisioning Management Server name. Using the virtual computer name can cause a failure to connect
SystemProvisioning.
78
2.7 Log Settings
The output settings of the debug log can be changed using the System Monitor - Performance Monitoring Services management
console. A log level, a maximum size of the log file and the log file generation count can be changed.
A log level can be specified between '0' and '4'. The higher a log level becomes, the more detailed the log becomes. When '0' is
set as a log level, no log is output. The default value of log level is '3'.
A log file size can be specified between 1 and 1024MB. The default value of log file size is 10MB. A debug log file is backed
up if the file size exceeds the specified value. The file can be backed up as much as the file generation counts.
The maximum file count can be specified between 2 and 1000. The default value of log file size is 10.
The procedure for configuring these settings is as follows:
1. Start the management console and connect to the management server. See "3.1.2 Starting the Management Console" for
information on how to start the management console.
2. Right-click the management server name from the navigation tree window in the main window of the management
console. A popup menu appears.
3. Click Set configuration.
The Configuration window appears.
4. Specify the debug log level, the file size and the file count on the Log tab.
5. Click OK.
79
Chapter 3 Basic Operations
3.1 Starting the Management Console and Connecting to the Management Server
3.1.1 About the Management Console
The management console is the user interface for the Performance Monitoring Service. Start the management
console and connect to the management server to configure the Performance Monitoring settings. The
management console can also graphically display performance data.
You can connect to multiple management servers from one management console. This allows for central
management of multiple management servers. Also, more than one management console can connect to the
same Performance Monitoring Service at the same time. See "3.1.5 Notes on Operating with More than One
Management Console" for more information.
3.1.2 Starting the Management Console
Start the management console to initiate the connection between the management console and the
management servers. The execution account of the management console must possess OS administrator
authority of the Monitoring terminal. The procedure for starting the management console is as follows:
1. Point to SigmaSystemCenter under the All Programs section of the Start menu, and then click System Monitor
Management Console.
2. When no management server is registered in the management console, enter the necessary items into the Management
Server window.
Enter the hostname/IP address of the management server into Host name, the port number used for
connecting to the management server into Port number, the connection account to the management server
into User Name and the password into Password.
80
If the management console program is blocked by Windows Firewall,
Windows Security Alert window may appear. See "2.4.2 Firewall" and
unblock the management console program.
When a management server is already registered in the management console, enter the necessary items into the Logon
window. The account is used for connecting all management server registered in the management console.
Enter the connection account to the management server into User Name and the password into Password.
3. Click OK to start the connection to the management server.
3.1.3 Connecting to the Management Server
Start the management console to begin a connection between the management servers and the management
console. The management servers that are already registered in the management console are now connected.
The user name and password specified at management console startup are used for connecting to the
management server. Only users that have OS administrator authority for the management server can log on to
the management server. To connect, the Performance Monitoring Service must be started on the management
server. See "2.1 Starting the Performance Monitoring Service" for information on how to start the Performance
Monitoring Service.
If no management server is registered in the management console, the Management Server window appears
when the management console loads. In the Management Server window, add a management server. See "2.2
Registering the Management Server" for information on how to add management server.
3.1.4 Specifying from the Navigation Tree Window
From the navigation tree window, the following actions are possible: • Connecting to/Disconnecting from the management server • Changing connection settings • Deleting management server registration
To connect to or disconnect from the management server, complete the following steps:
1. Right-click the management server name and then click Connection.
A popup menu appears.
2. Click Disconnect.
Note
81
To reconnect, complete the following steps:
1. Right-click the management server name, and then click Connection.
A popup menu appears.
2. Click Connect.
To change the connection settings to the management server, complete the following steps:
1. Right-click the management server name, and then click Connection.
A popup menu appears.
2. Click Set.
A Management Server window appears.
3. Change the settings, and then click OK.
To delete management server registration, complete the following steps:
1. Right-click the management server name, and click Delete management server.
A confirmation window appears.
2. Click Yes.
Verify that the management server was deleted from the navigation tree window.
3.1.5 Notes on Operating with More than One Management Console
With System Monitor - Performance Monitoring Services, more than one management console can connect to
the same Performance Monitoring Service at the same time.
The following settings are unique for each Performance Monitoring Service:
• Specifications for monitored machines
• Specifications for the Performance Indicators to be collected
• Specifications for threshold monitoring information
• Pausing and restarting the collecting of performance data
• Configuration settings
Caution: If these settings are changed on one management console, the changes are reflected on all
management consoles. The displayed performance status information will also change.
82
Also, setting changes may not take effect if similar changes are made from another management console at the
same time.
It is possible to start multiple management consoles on a single machine. However, in order for this to work
correctly, the port numbers for each management console should be set to a different value.
Performance Monitoring Service
Management console
Changing settings
Management console
Changing settings
Performance Monitoring Service
Management console
Management console
Displaying performance
status
Applying
Applying
Displaying performance
status
Changing settings
Applying
Management console
Displaying performance
status
83
3.2 The Main Window
3.2.1 The Structure of the Main Window
The main window is displayed when users start the management console. The System Monitor - Performance
Monitoring Services main window is shown below.
3.2.2 Menu Bar
Each menu in the main window and its options are explained below.
File menu
Exit
Closes the management console.
View menu
Tree Window
Shows or hides the navigation tree window.
Menu bar Toolbar
Navigation tree window
Status bar Graph display pane
Search view
Log window
84
Tree Window Search View
Shows or hides the Tree Window Search View.
Toolbar
Shows or hides the toolbar.
Status bar
Shows or hides the status bar.
Log Window
Shows or hides the Log window.
Graph menu
New
Displays the Graph Setting window. Select this menu item to create a new graph. See "Chapter 5
Performance Status Display and File Output" for more information on the Graph Setting window.
Set Display
Displays the Graph Setting window for the active performance graph window in the current graph display
pane. Select this menu item to change the settings for a graph that is being displayed. See "Chapter 5
Performance Status Display and File Output" for more information on the Graph Settings window.
Output File
Displays the Output Settings window. Select this menu item to output performance data displayed in the
graph window. See "5.3 File Output" for more information on the Output Settings window.
Option
Displays the Edit Performance Indicator window. Specify the Performance Indicator title to be displayed in
graphs. See "5.1.4 Defining a Performance Indicator Title" for more information on the Edit Performance
Indicator window.
Tool menu
Add Management Server
Displays the Management Server window. Select this option when adding a management server. See "2.2
Registering the Management Server" for more information on the Management Server window.
Change Connection Status of Management Servers
You can select Connect to all or Connect to Management Servers with error status. Select Connect to all to
connect to all management servers with error status or those that are disconnected. Select Connect to
Management Servers with error status to connect to all management servers with error status.
Change Data Collecting Status
You can select Restart All or Pause All. Specify restart/pause Performance Monitoring Service for all
connected management servers.
85
Apply SystemProvisioning Configuration to All
Configuration information on SystemProvisioning is applied to all connected management servers. Select
this menu item to apply configuration information immediately, regardless of periodic application
performed by the automatic applying function.
Set Management Console
Displays the Management Console window. See "2.4 Connection Settings to the Management Console" for
more information on the Management Console window.
Window menu
Cascade
Arranges the performance graphs so they overlap.
Tile Horizontally
Arranges the performance graphs horizontally.
Tile Vertically
Arranges the performance graphs vertically.
A list of performance graphs that are being displayed
Performance graphs that are being displayed are listed. The selected performance graph is activated.
Help menu
Version Information
Displays version information for System Monitor - Performance Monitoring Services.
3.2.3 The Toolbar
Each tool button in the main window is explained below.
Add Management Server tool button
Displays the Management Server window. Select when adding a management server. See "2.2 Registering
the Management Server" for more information on the Management Server window.
Restart Collecting Performance Data tool button
Restarts the Performance Monitoring Service for all connected management servers.
Pause Collecting Performance Data tool button
Pauses the Performance Monitoring Service for all connected management servers.
Apply SystemProvisioning Configuration to All tool button
Applies configuration information from SystemProvisioning to all the connecting management servers.
Select this tool button to apply configuration information immediately, regardless of periodic application
performed by the automatic applying function.
Create a Graph tool button
Displays the Graph Setting window. Select this tool button to create a new graph. See "Chapter 5
Performance Status Display and File Output" for more information on the Graph Setting window.
86
3.2.4 The Navigation Tree Window
Management servers and monitored machines are displayed in the navigation tree window. Operations for
management servers such as collecting data settings and adding monitored machines are performed from the
navigation tree window. See “3.2.5 Popup Menu for Management Server” for navigation tree window operations
for each management server.
3.2.5 Popup Menu for Management Server
Connection
Specifies connect to/disconnect from the management server. Set menu displays Management Server
window. See "2.2 Registering the Management Server" for more information on the Management Server
window.
Delete Management Server
Deletes management server registration.
Collecting Data
Select the Set Performance Indicator Settings menu or the Set Collecting Data Settings menu.
If the Set Performance Indicator Settings menu is selected, the Performance Indicator List window appears.
Custom performance indicator setting can be added or modified on the Performance Indicator List window.
If the Set Collecting Data Settings menu is selected, the Collecting Data Setting List window appears. You
can specify which node and performance indicator should be monitored on the Collecting Data Setting List
window.
See "4.2 Specifications for the Performance Data to be Collected" for more information about each window.
Set Threshold Monitor
Displays the Set Threshold Monitor window. See "6.1 Configuring Threshold Monitoring" for more
information on the Set Threshold Monitor window.
Change Data Collecting Status
Specifies restart/pause Performance Monitoring Service for the management servers. You can select
Restart or Pause.
Apply SystemProvisioning Configuration
Applies configuration information from SystemProvisioning to the management server. Select this menu to
Apply configuration information immediately, regardless of periodic application performed by the
automatic applying function.
87
Set Configuration
Displays the Configuration window. See "Chapter 2 Configuration Settings" for more information on the
Configuration window.
Add Group
Displays the Group Settings window. See "4.1.1 Specifications in the Navigation Tree Window" for more
information on the Group Settings window.
Display Graph
Creates a real-time graph using the simple graph display function. See "5.1.3 Simple Graph Display
Function" for more information on the Simple graph display function.
3.2.6 Search View
Select Tree Window Search View from the View menu to display search view. Enter search character strings to
the text box then click Search. When there is a node name partially matching to the search character string in
the navigation tree window, the corresponding node switches to select status. Click Search again to switch the
next corresponding node to select status. The search sequence is performed from the top of the navigation tree
to the bottom of the navigation tree. When there are no corresponding nodes, the select status on the navigation
tree window is not changed.
3.2.7 The Graph Display Pane
Performance graph windows are displayed in the graph display pane. See "Chapter 5 Performance Status
Display and File Output" for more information on the display content.
88
3.2.8 Log Window
Events that occur in Performance Monitoring Service and the management console are displayed in the Log
window. Right-click in the log window area, and then click Setting. The Log Settings window appears. You can
specify the contents to be displayed in the Log window in the Log Settings window. The line count to be
displayed ranges from zero to 1000 lines, which can be specified in one-line units.
If you want to copy a log, select the log on the log window, right-click on the log window area and click Copy
from the pop-up menu. You can save the log easily and quickly by copying and pasting the log into other file.
3.2.9 The Status Bar
The current time is displayed in the status bar.
89
Chapter 4 Collecting Performance Data
You can manage multiple management servers with System Monitor - Performance Monitoring Services. To do
this, you must first connect to the management server to start collecting performance data. See “3.1.3 Connecting
to the Management Server” for more information about connecting to the management server. Monitored
machines and performance data to be collected are specified for each node from the navigation tree window. See
“4.1 Specifications for Monitored Machines” for more information about specifying a monitored machine, and see
“4.2 Specifications for the Performance Data to be Collected” for more information about specifying performance
data.
4.1 Specifications for Monitored Machines
4.1.1 Specifications in the Navigation Tree Window
Additions, deletions and changes to monitored machines are performed from the navigation tree window of the
main window. Monitored machines are configured for each management server.
System Monitor - Performance Monitoring Services manages monitored machines in groups called "groups".
Performance status information can be displayed for the entire group, so you must register groups of machines
whose performances need to be monitored collectively as groups.
The following section explains how to register machines and groups.
90
(1) Initial State
When there are registered servers, the management server name and group are displayed in the
navigation tree window. The default name for the default group is "Group1".
(2) Adding, Changing and Deleting Machines
Adding Machines
The procedure for adding machines to a group is as follows:
1. Right-click the group name, and then click Add Machine.
The Machine Settings window appears.
2. Enter the host name or IP address for the machine in the Machine Name text box. (Required)
3. To set the IP address explicitly, enter the IP address in the IP Address text box. (Optional)
Note: If an IP address has been set up, the specified IP address is used to connect to this machine from the management
server. An IP address only can be set up if the Apply IP address information to machines when applying
SystemProvisioning Configuration box has not been checked in the Group Settings window. If the Apply IP address
information to machines when applying SystemProvisioning Configuration box has been checked, then the IP address
that was acquired from SystemProvisioning when SystemProvisioning configuration information was applied is set up
automatically. Windows and Linux support both IPv6 and IPv4 addresses. Other operating systems than Windows and
Linux support IPv4 addresses only.
The '\' character cannot be used in the Machine Name text box. Note
91
4. Select the appropriate OS from the OS name list.
Note: The OS Name list can only be set if the Apply SystemProvisioning Configuration box has not been checked in
the Group Settings window. If the Apply SystemProvisioning Configuration box has been checked, then the OS name
that was acquired from SystemProvisioning when SystemProvisioning configuration information was applied is set up
automatically.
5. For Weight, specify a weighting for the machine between 0.0 and 10000.0. (Optional)
Set the Weight for this machine if you want the weighted average value statistic to be used. For more information on
Weight, see "1.8.1 Statistics".
6. In the Memo field, enter in a memo or note to be displayed for the machine. (Optional)
Click the Connection tab and perform connection settings. When using connection settings of group, set the checkbox to
ON. When performing connection settings individually, set the checkbox to OFF and specify the account/password.
When the monitored machine OS is Windows, select Windows OS Account or SSH Account (only for script collection)
according to your purposes.
• Specify the account information for performance data collecting with performance counters in Windows OS
Account.
• Specify the account information for performance data collecting with scripts in SSH Account(only for script
collection).
For the SSH user account in Windows or when the monitored machine OS is Linux/KVM, select the protocol to be used
for connection and the authentication method. When public key authentication method is selected, specify the full path
name of the private key file stored in the Management Server. See "1.7 Connecting to Monitored Machines" for
applicable rules for connection settings to managed machines, general notes for settings and monitored machine side
settings.
92
<When Monitored Machine OS is Windows>
<When Monitored Machine OS is Linux or KVM>
93
<When Monitored Machine OS is VMware ESX/ESXi>
<When Monitored Machine OS is Citrix XenServer>
7. Click the OK button when settings are completed.
94
8. Verify that the machine has been added to navigation tree window.
Deleting Machines
The procedure for deleting machines is as follows:
1. Right-click the machine name and click Delete Machine.
A window appears to confirm that you want to delete the machine.
2. Click Yes.
3. Verify that the machine has been deleted from the navigation tree window.
The machine is deleted from the navigation tree window, but information will still be stored in the database (for
a storage period of one month). If a machine with the same machine name is added, the configuration will revert
to the former state (the data stored in the database will become valid).
Confirming or Modifying Machine Settings
The procedure for confirming or modifying machine settings is as follows:
1. Right-click the machine name and click Machine Settings.
The Machine Settings window appears.
2. Confirm the machine settings. Enter the new machine settings if necessary. The information on Summary tab in the Set
server window is displayed based on SystemProvisioning configuration information. See "1.10.2 Applying System
Configuration Information" for information on applying SystemProvisioning configuration information.
3. Click OK.
4. Verify that the new settings were applied if they were modified.
Moving Machines
The procedure for changing which group a machine belongs to is as follows:
1. Right-click the machine name and click Move machine.
The Move Machine window appears. It contains a list of groups.
2. Select the new group from the window.
3. Click OK.
4. Verify that the machine has been moved to the new group in the navigation tree window.
You can also drag and drop the machine name to a different group, but you cannot move a machine to the group of another
The settings applied from SystemProvisioning are grayed out on this window. These
settings cannot be modified. The settings on SystemProvisioning should be modified if
necessary.
Note
95
management server.
(3) Adding, Changing and Deleting Groups
Adding Groups
The procedure for adding groups is as follows:
1. Right-click the management server and click Add Group.
The Group Settings window appears.
2. Enter the name for the group in the Group Name field. (Required)
Note: The '\' character cannot be used in the Group Name field.
3. To apply configuration information for SystemProvisioning on to the group that is being set up, check the Apply
SystemProvisioning Configuration box and specify the path for the appropriate SystemProvisioning group/model in the
Path text box. To apply IP address information to subordinate machines when configurations are applied, check the
Apply IP address information to machines when applying SystemProvisioning Configuration box. (Optional)
4. In the Memo field, enter in a memo or note to be displayed for the group.
5. Click the Connection tab and perform connection settings. When using connection settings of the Configuration window,
set the checkbox to ON. When performing connection settings individually, set the checkbox to OFF and specify the
account/password. When the monitored machine OS is Linux is set to monitored machines, select the protocol to be used
for connection and the authentication method. When public key authentication method is selected, specify the path name
of the private key file stored in the management server. See “1.7 Connecting to Monitored Machines” for applicable
rules for connection settings to managed machines, general notes for settings and monitored machine side settings.
96
6. Click the OK button when settings are completed.
7. Verify that the group has been added to the navigation tree window.
Deleting Groups
The procedure for deleting groups is as follows:
1. Right-click the group and click Delete Group.
A window appears to confirm that you want to delete the group.
2. Click Yes.
3. Verify that the group has been deleted from the navigation tree window.
Performance data for groups that have been deleted becomes unusable. The performance status from before the deletion will not be displayed even if a group with the same name is added.
Note
97
Confirming or Modifying Group Settings
The procedure for confirming or modifying group settings is as follows:
1. Right-click the group name and click Group Settings.
The Group Settings window appears.
2. Confirm the group settings.
If the performance data collecting setting is enabled on the Performance Monitor settings about the corresponding
group/model on SystemProvisioning, an access account can be configured there. An access account can be configured on
the Connection settings for each node on System Monitor - Performance Monitoring Service as well. You can select
which account is used as an access account for this group.
In order to use the access account specified on SystemProvisioning, set the Use connection setting of
SystemProvisioning box to ON.
When the application of SystemProvisioning configuration information is performed, the access account which is specified on SystemProvisioning is applied to the group settings of System Monitor - Performance Monitoring Service.
Note
98
In order to use the access account specified on System Monitor - Performance Monitoring Service, set the Use
connection setting of SystemProvisioning box to OFF and set the connection settings.
Enter the other new group settings if necessary.
3. Click OK.
4. Verify that the new group settings were applied if they were modified.
The settings applied from SystemProvisioning are grayed out on this window. These
settings cannot be modified. The settings on SystemProvisioning should be modified if
necessary.
Note
The '\' character cannot be used in the Group Name text box. Note
99
For some system configurations, there are cases where it is not possible to
connect to monitored machines using the IP addresses acquired from
SystemProvisioning configuration information. If a monitored server
produces an error, disable Apply the IP address information function and set
the IP address manually using the Set server window, DNS or the "hosts"
file.
4.1.2 How to Apply SystemProvisioning Configuration Information
SystemProvisioning configuration information can be applied to System Monitor - Performance Monitoring
Services. Information can be applied manually or automatically. With the automatic applying function, System
Monitor - Performance Monitoring Services communicates with SystemProvisioning at regular intervals, and
automatically applies any changes to its configuration information.
Set the following attributes beforehand for System Monitor - Performance Monitoring Services groups for which
SystemProvisioning configuration information is applied (these settings can be made from the Group Settings
window):
1) Whether or not to apply SystemProvisioning configurations
2) The corresponding SystemProvisioning full path to the group/model
3) Whether or not to acquire IP address information
Configuration information, such as machine names and machine states (normal or error), is applied to the
machines in the specified SystemProvisioning group/model.
(1) Applying SystemProvisioning Configurations Manually
The procedure for applying SystemProvisioning configurations is as follows:
1. Select Apply SystemProvisioning Configuration to All from the Tools menu, or click on the Apply
SystemProvisioning Configuration to All tool button.
A window appears to confirm whether the SystemProvisioning configuration should be applied.
2. Click Yes.
3. Verify that the configuration has changed in the navigation tree window.
(2) Applying SystemProvisioning Configurations Automatically
The setup procedure to apply SystemProvisioning configurations automatically is as follows:
1. Right-click the management server in the navigation tree window and click Set Configuration.
The Configuration window appears.
2. In the SystemProvisioning tab, set the required items. See “2.6 SystemProvisioning Connection Settings” for
more information.
3. Click OK.
Every time SystemProvisioning is polled, any configuration changes that have been made are applied in the navigation
tree.
Note
100
4.2 Specifications for the Performance Data to be Collected
4.2.1 How to Specify Performance Indicator Settings
System Monitor - Performance Monitoring Services can collect performance data for the custom performance
indicator as well as the built-in performance indicator.
To specify performance indicator, complete the following steps:
1. Right-click the management server.
2. Select Collecting Data menu.
3. Click Set Performance Indicator Settings menu.
The Performance Indicator List window appears.
The performance indicators written in "1.3.2 Built-in Performance Indicator" are registered as default settings.
When you want to create a new performance counter except the built-in performance counter for the Windows
machines and the ESX/ESXi virtual machine servers and virtual machines, or when you create a new scripted
performance indicator for Windows or Linux machines, you can create it as the custom performance indicator.
101
Adding Performance Indicators
The procedure for adding new Custom Performance Indicators is as follows:
Performance Counters
1. Click the Add Performance Counter button on the Performance Indicator List dialog box.
2. The Performance Indicator Settings dialog box appears. It displays all object nodes monitored by System Monitor -
Performance Monitoring Services in the tree. The management server/node is selected and performance counter setting is
disabled by default.
102
3. Select the source object node from the Object Node list to collect the information from the operating system on the
selected node, and to set up performance counters.
4. Specify Data Provider.
• For VMware ESX/ESXi virtual machine servers, you can specify only "VMware Esx Host".
The "VMware Esx Host" data provider obtains only performance data which a virtual machine server can obtain from
VMware ESX/ESXi.
• For VMware ESX/ESXi virtual machines running Windows, You can specify both "VMware Esx Guest" and
"Windows". For VMware ESX/ESXi virtual machines running other operating systems than Windows, you can specify
"VMware Esx Guest" only.
The "VMware Esx Guest" data provider obtains performance counters which a virtual machine can obtain from
VMware ESX/ESXi.
The "Windows" data provider obtains performance counters which Windows can obtain.
• For other machines running Windows than VMware ESX/ESXi virtual machines, you can specify "Windows" only.
• In all other cases, the performance counter is set to "Disable" and cannot be configured.
103
5. Click Get Counter. The list of performance counter information which can be configured on the selected object node is
obtained and displays each performance counter information in "Performance Counter (Windows) / Metric Group
(VMware ESX/ESXi)", "Counter", and "Instance (Windows) / Object (VMware ESX/ESXi)".
6. Select the resource you want to use from Performance Counter (Windows) / Metric Group (VMware ESX/ESXi), the
counter from Counter, the instance from Instance (Windows) / Object (VMware ESX/ESXi) respectively.
You can select multiple counters, instances (Windows), and objects (VMware ESX/ESXi) and add them in bulk. When
you want to select or cancel the selection of multiple counters, instances (Windows), and objects (VMware ESX/ESXi)
click Select All or Clear All.
7. Multiple performance indicators in performance counters (Windows) or metric groups (VMware ESX/ESXi) can be
configured by repeating the actions in step 6. When you are finished configuring the settings, click Close.
Note
It takes some time to initially obtain the performance counter information
from monitored machines running Windows Vista, Windows 7, Windows
Server 2008 or Windows Server 2008 R2. This temporally causes lock
condition in the System Monitor - Performance Monitoring Services console
or might cause the time-out. If the process timed-out, click Get Counter to
retry the process.
104
8. Click Apply All in the Performance Indicator List window to apply these settings to all the connected management
servers.
User Scripts
1. Click Add Executing Script button on the Performance Indicator List dialog box.
2. The Performance Indicator Settings dialog box appears.
3. Specify a performance indicator:
・ Type the title of the performance indicator. (Required)
・ Type the execution string into Executing String (Required)
・ Type the description of the performance indicator into Description. (Optional)
4. Confirm that the specified performance indicator is added to the performance indicator list.
5. Multiple performance indicators can be configured by repeating the actions in step 3. When you are finished configuring
the settings, click Close.
6. Click Apply All in the Performance Indicator List window to apply these settings to all the connected management
servers.
Collecting performance data is not started only by adding the performance indicator
settings to the performance indicator list. For data collection settings, see "4.2.2
How to Specify Collecting Data Settings".
Note
If Apply All is executed, the settings which had been already specified on other
management servers connecting with the management console will be overwritten.
Note
105
Deleting Performance Indicators
The procedure for deleting Performance Indicators is as follows:
1. Display the Performance Indicator List window.
2. Select the Performance Indicator item to be deleted from the Performance Indicator list.
You cannot delete the built-in performance indicator.
3. Click Remove.
4. Verify that the Performance Indicator item was deleted from the Performance Indicator list.
Changing Performance Indicators
The procedure for changing Performance Indicators is as follows:
Performance counters
1. Display the Performance Indicator List window.
2. Select the Performance Indicator item to be changed from the Performance Indicator list.
Built-in performance indicators cannot be modified.
3. Click Modify.
4. The Performance Indicator Settings window appears.
5. Modify the Performance Indicator settings, and then click OK.
6. Click Apply All in the Performance Indicator List window to apply these settings to all the connected management
servers.
Performance data that has been deleted becomes unusable. The performance status
from before the deletion will not be displayed even if the same settings are made
again.
Note
106
User scripts
1. Open the Performance Indicator List dialog box.
2. Select the custom performance indicator that you want to modify from the Performance Indicator List. Built-in
performance indicators cannot be modified.
3. Click Modify.
4. The Performance Indicator Setting of the performance indicator to be modified appears.
5. Modify the performance indicator settings, and then click OK.
Click Apply All in the Performance Indicator List window to apply these settings to all the connected
management servers.
If Apply All is executed, the settings which had been already specified on other
management servers connecting with the management console will be overwritten.
Note
107
4.2.2 How to Specify Collecting Data Settings
System Monitor - Performance Monitoring Services periodically collects performance data for monitored
machines. The Performance Indicators that are collected can be specified for each management server, for each
group or for each machine. The settings for collecting performance data from the monitored machine can be
specified on the Collecting Data Setting List window by assigning the collecting data settings to the object node.
The procedure for specifying settings for collecting performance data is as follows:
1. In the navigation tree window, right-click the management server, group or monitored machine that you want to configure
the settings, and select Collecting Data.
2. Click Set Collecting Data Settings menu.
The Collecting Data Setting List window appears.
3. Select a node in the Object node list and the current settings for the selected node can be seen in the Collecting data
setting list. It is set as inheriting from the parent node's settings by default.
108
4. To modify the settings for the selected node, clear the Inherit from parent settings box.
5. To add a collecting data setting to the list for the selected node, click Add. To remove a collecting data setting from the
list for the selected node, select the setting to be removed and click Remove. To modify a collecting data setting, select
the setting to be modified and click Modify.
6. Add or modify the collecting data setting performing the following procedure.
If the collecting data setting which is target of threshold monitoring is deleted, the
threshold monitoring is stopped. But the corresponding performance monitor
settings are not deleted automatically. If necessary, delete these performance
monitor settings manually on the Set Threshold Monitor window.
See "6.1.1 How to set up threshold monitoring" for details on the Set Threshold
Monitor window.
Note
109
Adding the collecting data setting
1. The Collecting Data Settings window appears when Add is clicked.
2. Specify the collecting data settings to be added.
• Select a resource name of the collecting data setting to be added or modified in the Resource list. To specify the
custom performance indicator, select "Other".
• Select a performance indicator name of the collecting data setting to be added or modified in the Performance
indicator list. You can select multiple performance indicators and add them in bulk. When you want to select or
cancel the selection of multiple performance indicators, click Select All or Clear All. When the specified resource
has performance indicator too many, you can use the search feature to find out them easily. Type a search string into
the text box and click Search, then the performance indicators that contain the search string in their title are listed in
the Performance indicator list.
• Select an interval of the collecting data setting to be added or modified in the Collecting interval list.
3. Click Add.
110
Modifying the collecting data setting
1. The Collecting Data Settings window appears when Modify is clicked.
2. Specify the collecting data settings to be modified.
• Select a resource name of the collecting data setting to be modified in the Resource list. To specify the custom
performance indicator, select "Other".
• Select a performance indicator name of the collecting data setting to be modified in the Performance indicator list.
• Select an interval of the collecting data setting to be modified in the Collecting interval list.
3. Click OK.
111
The function of collecting performance data from Windows machine will not
perform when the specified collecting interval is a day or longer. It is
necessary for collecting performance data from Windows machine that a
shorter interval than a day is specified as collecting interval.
7. Verify that the new settings were applied to the list on the Collecting data settings List window.
8. Click OK.
The following four performance indicators are the default settings:
Performance Indicator Location Collecting Interval
CPU Usage (%) Root 1 minute
Disk Transfer Rate (Bytes/sec) Root 1 minute
Disk Space (MB) Root 1 minute
Physical Memory Space (MB) Root 1 minute
Note
112
4.2.3 Collecting Data Settings Automatically Applied by SystemProvisioning Linkage
The system configuration information and collecting data settings (Monitoring Profile) specified on SystemProvisioning can
be applied automatically to System Monitor - Performance Monitoring Services. For applying the settings on
SystemProvisioning to System Monitor - Performance Monitoring Services, it is necessary to specify the monitoring profile
and enable the performance data collection settings on the group/model property.
Applied collecting data settings from SystemProvisioning to System Monitor - Performance Monitoring Services are listed in
blue in the Collecting Data Setting List window.
It is impossible to delete or modify the applied collecting data settings, but it is possible to add other collecting data settings to
the list for the automatically applied node.
113
4.3 Pausing and Restarting the Collecting of Performance Data
System Monitor - Performance Monitoring Services can pause and restart the collecting of the performance data.
To pause all performance data collection on all monitored machines, select Change Data Collecting Status and
then click Pause All from the Tool menu. To restart the collection of performance data, select Change Data
Collecting Status and then click Restart All from the Tool menu.
Performance data collecting can also be paused for the management server, the group or the monitored machine.
To do this, complete the following steps:
1. In the navigation tree window, right-click the management server, group or monitored machine that you want to pause,
and click Change Data Collecting Status
2. In the submenu that appears, click Pause.
To restart the service, complete the following steps:
1. In the navigation tree window, right-click the management server, group or monitored machine that you want to restart ,
and click Change Data Collecting Status.
2. In the submenu that appears, click Restart.
114
4.4 Checking the Collecting Status for Performance Data
The collecting status for performance data is displayed in the navigation tree window of the main window. The
collecting status can be checked by looking at the icons in the tree display that shows the collecting status (the
status can be any of the following: "collecting", "paused" or "error") for each management server, group or
monitored machines.
The meaning of each icon is as follows:
(1) Management Servers
Connected to the management server, data is being collected normally for all
machines under the management server
Disconnected from the management server
Connection Error
Connected to the management server, data collecting for all machines under
the management server has been paused
Connected to the management server, data collecting for a machine or
machines under the management server has been paused
Connected to the management server, data collecting failed for a machine or
machines under the management server
Connected to the management server, data is being collected normally for all
machines under the management server and an over-threshold status has
occurred
Connected to the management server, data collecting for a machine or
machines under the management server has been paused and an
over-threshold status has occurred
Connected to the management server, data collecting failed for a machine or
machines under the management server and an over-threshold status has
occurred
115
(2) Monitored Machines
Data is being collected
Data collecting has been paused
Collecting failed for some Performance Indicator items
(performance indicator error status)
The machine cannot be accessed or collecting failed for all Performance
Indicator items (access error status)
Data is being collected, and an over-threshold status (warning) has occurred
Data is being collected, and an over-threshold status (critical) has occurred
Collecting failed for some Performance Indicator items, and an over-threshold
status (warning) has occurred
Collecting failed for some Performance Indicator items, and an over-threshold
status (critical) has occurred
116
(3) Groups
Data is being collected normally for all machine in the group
Data collecting for the group has been paused
Data collecting for a machine or machines in the group has been paused
Collecting failed for some Performance Indicator items for a machine or machines
in the group
A machine in the group cannot be accessed or collecting failed for all Performance
Indicator for a machine or machines in the group
Data is being collected normally for all machines in the group and an
over-threshold status (warning) has occurred
Data is being collected normally for all machines in the group and an
over-threshold status (critical) has occurred
Data collecting for a machine or machines in the group has been paused and an
over-threshold status (warning) has occurred
Data collecting for a machine or machines in the group has been paused and an
over-threshold status (critical) has occurred
Collecting failed for some Performance Indicator items for a machine or machines
in the group and an over-threshold status (warning) has occurred
Collecting failed for some Performance Indicator items for a machine or machines
in the group and an over-threshold status (critical) has occurred
One or more machines in the group cannot be accessed or collecting failed for all
Performance Indicators for a machine in the group, and an over-threshold status
(warning) has occurred
One or more machines in the group cannot be accessed or collecting failed for all
Performance Indicators for a machine in the group, and an over-threshold status
(critical) has occurred
117
Chapter 5 Performance Status Display and File Output
5.1 Displaying Performance Graphs
System Monitor - Performance Monitoring Services displays performance data for the system graphically.
Displaying performance data graphically enables trends to be identified intuitively, and makes it easier to
analyze the operational status of the system.
Graphs of Performance Indicators that are currently being collected can be displayed for each management
server, for each group or for each individual machine. See “5.1.1 How to Display Performance Graphs” for more
information about displaying the Performance Indicator graphs. Also, see “5.1.3 Simple Graph Display Function”
for information on the use of the simple graph display, which uses default values.
See “5.3 File Output” for methods to output performance data displayed in a graph to CSV or other files.
5.1.1 How to Display Performance Graphs
Performance graphs are displayed based on performance data. The Performance Indicators must be configured
prior to use.
Specify graph displays using the following procedure:
118
1. Select New from the Graph menu or click the Create a Graph icon.
The Graph Setting window appears.
2. Enter a graph title in the Graph Title box. A title is automatically generated if the Generate Graph Title Automatically
box is checked. (Required)
3. Select which contents to display in the graph by using the Display Object tab. Articles for Select Item for Comparison,
Nodes, Performance Indicators and Statistics are displayed in the Display Object tab. You can select multiple items for
whichever choice you pick in the Item for Comparison area, and select one item in the other articles. For example, if you
select Nodes in the Item for Comparison area, you can select multiple nodes in the Nodes article and one Performance
Indicator in the Performance Indicators article and one statistic in the Statistics article. At least one item must be
selected. (Required)
4. To add nodes to the Nodes column, click Add.
A Browse nodes window appears.
5. Select the nodes you want to add, and click OK.
119
6. If the Performance Indicator title for the management server doesn't appear in the Performance Indicators column, click
Cancel in the Graph Setting window. You must define a Performance Indicator titles in the management console
beforehand. See “5.1.4 Defining a Performance Indicator Title” for information on how to add a Performance Indicator
title.
7. Click the Horizontal Axis tab, and make specifications relating to the time axis.
The settings that are made here depend on which display period was specified (Historical or Realtime).
120
For Historical graphs, make the following specifications:
• Display period
Specify the Start time and Period. Start time specifies when the graph will start displaying data, and
Period specifies the period of time of which the data is displayed in the graph.
Period can be specified between one and 1000 and in the following units: minutes, hours, days, weeks,
months, or years. You can specify Start time to start before the current time by checking the Start time is
specified period before current time box.
• Plotting interval
Specify individual settings or automatic settings. For individual settings, the following time periods can
be specified: 1 minute, 5 minutes, 30 minutes, 1 hour, 4 hours, 1 day, 1 week, or 1 month.
• Gridlines
Select automatic settings or individual settings for the gridline interval. For individual settings, the
Major and Minor gridline intervals can be specified between one and 1000 and in the following units:
minutes, hours, days, weeks, months, or years.
For Realtime graphs, make the following specifications:
• Display period: The display period can be specified between 1 and 1000. You can specify "minutes",
"hours", "days", "weeks", "months", or "years" for the units.
• Polling interval for updating data: You can specify "1 second", “5 seconds”, “20 seconds”, “1 minute”, “5
minutes”, “20 minutes”, “1 hour”, “4 hours”, or “1 day” for the graph update interval
• Plotting interval
Specify individual settings or automatic settings. For individual settings, the following time periods can
be specified: 1 minute, 5 minutes, 30 minutes, 1 hour, 4 hours, 1 day, 1 week, or 1 month.
• Gridlines: Select either automatic settings or individual settings for the gridline interval. For individual
settings, the Major and Minor gridline intervals can be specified between 1 and 1000. You can specify
"minutes", "hours", "days", "weeks", "months", or "years" as the unit for the gridline interval.
8. Use the Vertical Axis tab on the Graph Setting window to change graph vertical axis values from their default setting.
The following specifications can be made:
• Display range
Specify the upper- and lower-bounds between zero and 1000. The default upper-bound is 100, and the
default lower-bound is zero.
• Gridlines
Select either automatic settings or individual settings for the graduation interval. For individual settings,
uncheck the Automatically set gridlines display interval box and then specify the Major and Minor
gridline intervals. The gridline intervals can be specified between one and 1000.
• Scale
Select either automatic settings or individual settings for the Performance Indicator in the list box. The
default behavior is for the scale to be set automatically. For individual settings, check the Set the scale
121
box and select a scale for each Performance Indicator item as 10 to the nth power (where n is an integer
between -10 and 10). The data values are multiplied by the specified scale. The scale should be set so that
the data values fit within the vertical axis of the graph.
122
9. Use the Color and Line tab to modify the graph color and line settings from their default values.
The following specifications can be made:
• Graph lines
Select the graph feature that you want to modify from the Object list, and then specify Color, Line type
and Width.
• Graph construction lines
Specify Color, Line type (only for gridlines) and Width for the Vertical axis, Horizontal axis, Graph frame,
and Major and Minor gridlines.
• Background
Select the graph feature that you want to modify from the Object list. You can specify the colors for the
following: the background of plotting area; the background of graph; the upper-limit critical area; the
upper-limit warning area; the lower-limit critical area; and the lower-limit warning area.
123
10. Use the Threshold Display tab to specify graph thresholds.
11. Check the Display threshold area box.
124
12. Click Browse.
The Browse threshold definition window appears.
13. Select a threshold definition from the Threshold definition list, and then click OK to apply the threshold contents to the
Threshold Display tab. You can also manually specify each value for displaying threshold areas. To do so, you need to
specify Performance Indicator title for using scale too.
14. After making these specifications, click OK.
15. Verify that your changes are reflected in the performance graph that is displayed in the graph display pane.
125
5.1.2 Graph Window
Graph Window displayed on Graph Display Pane is explained below.
Graph
• Line Graph
Target data is displayed in line graph.
• Vertical Axis
The vertical axis range of the graph is configured in Vertical Axis tab on Graph Setting Window.
• Horizontal Axis
The horizontal axis range of the graph is configured in Horizontal Axis tab on Graph Setting Window.
• Threshold area
The threshold areas on the graph are configured in the Threshold Display tab on Graph Setting Window and
the colors of them are configured in Color and Line tab on Graph Setting Window.
Graph Summary List
• Node Name / Performance Indicator / Statistics
Target object names displaying as a graph are displayed. The settings in Display Object tab on Graph Setting
Window are displayed.
• Color / Line Type
Colors and line types used for displaying a graph are displayed. The settings in Color and Line tab on Graph
Setting Window are displayed.
Graph Summary List
Graph
126
Scale is updated in real time when automatic scale setting is enabled. Shape
of the graph will be changed when the scale is updated automatically.
• Latest Value / Maximum Value / Average Value / Minimum Value
Latest value, maximum value, average value and minimum value of the graph line currently shown on Graph
Window are displayed.
• Scale
Scale used to display the graph. Raw data values multiplied by the value of the scale are displayed in the
graph. The specified setting is displayed when the scale is specified manually in Vertical Axis tab on Graph
Setting Window. When automatical scale setting is enabled, an adjusted scale to display a whole graph within
a default vertical display range of 0 to 100 is displayed.
5.1.3 Simple Graph Display Function
You can easily display real time graphs with default values. The following graph types use default values to
quickly display graphs without having to configure specific information. The graph types and their default
values are listed below.
(1) Graph Display from Management Server Node
1. Right-click the management server name from the navigation tree window, and then click Display Graph.
2. Click Groups, Performance Indicators, or Statistics to display the corresponding graph. The following
default values are used.
Menu Node Performance Indicator Statistics Threshold
Groups All groups Top Performance
Indicator from Edit
Performance Indicator
window
Average None
Performance
Indicators
Management
server
All Performance
Indicators applied on
Collecting Data Setting
List window
Average None
Statistics Management
server
Top Performance
Indicator from Edit
Performance Indicator
window
All
statistics
except
Summation
None
Note
127
(2) Graph Display from Group Node
1. Right-click the group name from the navigation tree window, and then click Display Graph.
2. Click Machines, Performance Indicators, or Statistics to display the corresponding graph. The following
default values are used.
Menu Node Performance Indicator Statistics Threshold
Machines All machines
belonging to the
group
Top Performance
Indicator from Edit
Performance Indicator
window
Average None
Performance
Indicators
Group All Performance
Indicators applied on
Collecting Data Setting
List window
Average None
Statistics Group Top Performance
Indicator from Edit
Performance Indicator
window
All
statistics
except
Summation
None
(3) Graph Display from Machine Node
1. Right-click the machine name from the navigation tree window, and then click Display Graph.
2. Click Performance Indicators or VMServer Load Condition to display the corresponding graph. The
following default values are used.
Menu Node Performance Indicator Statistics Threshold
Performance
Indicators
Machine All Performance
Indicators applied on
Collecting Data Setting
List window
Average None
VMServer
Load
Condition
Virtual machine
server and
virtual machines
running on it
Top Performance
Indicator from Edit
Performance Indicator
window
Average Bound value of the VM
Optimized Placement
settings on
SystemProvisioning
The VMServer Load Condition menu is available in the following conditions:
• The selected machine is applied from SystemProvisioning configuration information.
• The selected machine is a virtual machine server or a virtual machine.
128
5.1.4 Defining a Performance Indicator Title
Performance Indicator titles must be defined in the management console beforehand to be displayed in a graph.
Performance Indicator titles are defined from the Edit Performance Indicator window. You can also add
Performance Indicator titles by clicking Apply All from the Collecting Data Settings window.
The following procedure is the method for defining Performance Indicator titles from the Edit Performance
Indicator window.
(1) To display the Edit Performance Indicator window, select Options from the Graph menu and then click
Performance Indicator.
The Edit Performance Indicator window appears.
(2) Initial state
The following four types of Performance Indicators are listed by default.
• CPU Usage (%)
• Disk Transfer Rate (Bytes/sec)
• Disk Space (MB)
• Physical Memory Space (MB)
129
(3) Add a Performance Indicator Title:
1. Click Add from the Edit Performance Indicator window.
2. The Browse Performance Indicator window appears.
3. Select the management server, and specify the Performance Indicator title to be added to the graph.
4. Click OK.
5. Verify that the Performance Indicator title has been added in the Edit Performance Indicator window.
6. On the Edit Performance Indicator window, click OK to add the Performance Indicator title.
(4) Delete a Performance Indicator Title
1. Select the Performance Indicator title from the Edit Performance Indicator window, and then click Remove.
2. Verify that the Performance Indicator title has been deleted from the Performance Indicator list.
3. On the Edit Performance Indicator window, click OK to delete the Performance Indicator title.
5.1.5 Changing Performance Indicator Graphs
The Performance Indicator graph settings can be changed using the following procedure:
1. Select the Performance Indicator graph that you want to modify. When you click the graph, the title bar will
turn blue.
2. Right-click the plotting area of the graph and select Set Display or select Set Display from the Graph menu.
The Graph Setting window appears.
3. The specification method for the Graph Setting window is the same as for "5.1.1 How to Display
Performance Graphs".
4. When you have finished modifying, click OK.
5. Verify the performance graph has changed.
130
5.2 Saving Performance Status Display Specifications
When the management console closes, settings relating to performance status display are saved, and these
settings are inherited next time the management console starts. The settings that are saved are as follows:
• Node comparison display versus Performance Indicator comparison display
• Real-time display versus specified period display (and the display period)
• Statistics to be displayed on the graph and graph display color.
• Show/hide settings for each Performance Indicator item
• Vertical axis display settings
• Plot interval settings
• Time axis display settings
• The display size and arrangement of the management console window
The settings relating performance status display are unique to each management server. If multiple management
consoles are started on a single management server, the settings for the last management console to be closed is
used the next time a management console is started.
If multiple management consoles are closed at the same time, the following error message may be displayed
during the termination processing. The settings for management consoles for which this error message is
displayed will not be saved.
The process cannot access the file “******” because it is being used by another process.
Management Server
Management console
Management console
File for saving display settings
When the management
console closes, performance status display settings are
updated
131
5.3 File Output
Outputs performance graph data displayed. It can be used with Microsoft Excel, and so forth, as the output file
formats are CSV and tab delineated text files.
5.3.1 File Output Methods
You must display performance data of output target beforehand on the performance graph window before
outputting files. See “5.1.1 How to Display Performance Graphs” for displaying performance graphs.
Output files in the following steps.
1. Set the performance graph of output target to active then click Output File in the Graph menu or
right-click in the plot area of the target performance graph, and then select Output File to display the
output settings window.
2. Specify filename
Filename is automatically set when the output settings window is displayed. To change the filename, direct
key input to the filename field or click the Browse button to select the output destination on the output file
window.
When the filename is changed to the relative path, the file is output to the relative path from the desktop
path of the current user.
Filenames automatically set consist of the following two parts.
[Path] + [Filename]
Path: The desktop path of the current user is set when the output settings window is displayed for the
first time. The path specified last time is set from the second time onward.
Filename: Set in a format shown below;
sysmon[Start Time]_[Span].[File Extension]
[Start Time] is the graph start time, and [Span] is the specified value of the graph display time span.
For [File Extension], “csv” or “txt” is set according to the file type.
132
3. Set file type.
When file type is changed, filename file extension, column separator and character encode format of the file to
be output are changed. See “1.8.5 Output Graph Data to File” for file type, column separator and character
encode format compatibility.
4. Set [Output the data in same line if time is same] checkbox.
Setting the checkbox to ON to enable specification outputs to same line as same time data when there are
differences in seconds units in multiple performance data time. The time to output to the file is formatted as
shown below at this time.
When less than 30 seconds, round down seconds.
When more than 30 seconds, round up seconds.
To output at exact time where performance data is collected, set the checkbox to OFF.
Specification is enabled while the checkbox is ON at preset.
5. Click the OK button to output the file with the specified filename. When a file already exists, an
overwrite confirmation window is displayed. When no problems exist when overwriting an existing file,
click the OK button.
133
Chapter 6 Threshold Monitoring and Notification
6.1 Configuring Threshold Monitoring
System Monitor - Performance Monitoring Services can detect and report load status events on monitored
machines by monitoring specified Performance Indicator thresholds Performance fault events can also be
notified to SystemProvisioning.
Thresholds can be set up for active Performance Indicators for both groups and individual machines.
6.1.1 How to set up threshold monitoring
Information for threshold monitoring is based on the set Performance Indicators that are collected. The
Performance Indicator settings must be set up beforehand.
Specify threshold monitoring settings using the following procedure:
1. Right-click the management server name in the navigation tree window, or the group name or machine name to be
monitored with the threshold. The pop-up menu appears. Click Set Threshold Monitor to display the Set Threshold
Monitor dialog box.
134
2. Configure combinations of the Object node and Threshold method as threshold monitoring items in the Performance
monitor list.
3. Specify which node to monitor. The nodes that can be specified as monitoring objects are displayed as a tree, depending
on the monitoring unit. Select the node from the tree. You can select one of the following three monitoring units: Monitor
machine, Monitor group or Monitor all machines in group. See "1.9 Threshold Monitoring and the Notification
Function" for more information on monitoring units.
4. Specify the threshold method.
5. To create a new threshold method, click New in the Threshold method section.
The Threshold Method Settings window appears. In the Threshold Method Settings window, configure the threshold
for one Performance Indicator item.
6. To modify an existing threshold method, select the threshold method from the list box, and then click Modify.
The Threshold Method Settings window appears. In the Threshold Method Settings window, configure the threshold
for one Performance Indicator item.
7. To delete an existing threshold method, select the threshold method from the list box, and then click Remove.
Specify the following items in the Threshold Method Settings window:
• Threshold method name
Enter the name of the threshold method in the Threshold method name text box. A threshold method
name can be generated automatically by clicking Generate Method Name. (Required)
135
• Performance Indicator and statistic
Select the Performance Indicator item and the statistics for the threshold to be defined from the
Performance Indicator and Statistics combo boxes. Statistics are only valid if the monitored object is a
group. (Required)
• Threshold and notification settings
Switch to the Upper Threshold, Lower Threshold and Notification Settings tabs and make further
settings.
8. When the Upper Threshold tab is selected, settings for upper-limit threshold values are displayed.
The following settings can be configured:
To specify upper-limit critical value monitoring, check the Enable upper-limit critical value monitor box.
When this box is checked, upper-limit critical value monitoring specifications can be made.
The title of the threshold method obtained from SystemProvisioning and
then registered begins with [VMServer] or [SystemProvisioning]. To make
the names of the threshold method unique, prevent the titles of manually
registered threshold method from beginning with [VMServer] or
[SystemProvisioning].
Note
136
• Threshold: Specify the threshold for upper-limit abnormalities. Specify a value between 0 and
3.402823e+38.
• Notification to SystemProvisioning: Check this box to notify SystemProvisioning when the upper-limit
warning value has been exceeded. When notifying SystemProvisioning, select notification types from the
Over and At recovery boxes.
To specify upper-limit warning value monitoring, check the Enable upper-limit warning value monitor box.
When this box is checked, upper-limit warning value monitoring specifications can be made.
• Threshold: Specify the threshold for upper-limit warnings. Specify a value between 0 and 3.402823e+38.
• Notification to SystemProvisioning: Check this box to notify SystemProvisioning when the upper-limit
warning value has been exceeded. When notifying SystemProvisioning, select notification types from the
Over and At recovery boxes.
137
9. When the Lower Threshold tab is selected, settings items for lower limit threshold values are displayed.
Specify the following items here:
To specify lower-limit critical value monitoring, check the Enable lower-limit critical value monitor box. When
this box is checked, lower-limit critical value monitoring specifications can be made.
• Threshold: Specify the threshold for lower-limit abnormalities. Specify a value between 0 and 3.402823e+38.
• Notification to SystemProvisioning: Check this box to notify SystemProvisioning when the lower-limit
critical value has been exceeded. When notifying SystemProvisioning, select the notification types from the
Over and At recovery boxes.
To specify lower-limit warning value monitoring, check the Enable lower-limit warning value monitor box.
When this box is checked, lower-limit warning value monitoring specifications can be made.
• Threshold: Specify the threshold for lower-limit warnings. Specify a value between 0 and 3.402823e+38.
• Notification to SystemProvisioning: Check this box to notify SystemProvisioning when the lower-limit
warning value has been exceeded. When notifying SystemProvisioning, select the notification types from the
Over and At recovery boxes.
138
10. When the Notification Settings tab is selected, settings items relating to notification are displayed.
Specify the following items here:
• Notification conditions
Set up notification conditions in terms of the number of times that over-threshold is detected during a
sequence of performance data checks. The performance data check interval is the performance data
collecting interval that was specified in the collecting data setting. By default, a notification is made if
over-threshold is detected ten times out of ten checks. The number of checks and the number of
over-threshold times can be set between one and 1000.
• If you want to send another notification if the performance status does not recover from over-threshold
status, then check the in case of no recovery from the over-threshold status box. If this box is checked, then
repeat notifications are enabled. Specify the timing for repeat notifications in terms of the number of checks
since the previous notification. The number of checks can be specified between one and 1000.
11. After you make specifications for the threshold method settings, click OK, and then verify that the settings have been
applied in the Threshold method list box in the Set Threshold Monitor window.
139
12. Select the item you want to configure as a threshold monitoring setting from the Object node or the Threshold method,
and then click Add to add the threshold monitoring item to the Performance monitor list. Multiple threshold monitoring
settings can be configured.
13. To delete an existing threshold monitoring setting, select from the Performance monitor list, and then click Remove to
the right of the list.
6.1.2 How to display over-threshold areas on graphs
The configure threshold method information can be displayed in the background of graphs as over-threshold
areas. In the Threshold Display tab of the Graph Setting window, check the Display thresholds area box and
then select the threshold method information to be displayed. The background color for over-threshold areas is
configured in the Color and Line tab of the Graph Setting window using the Background of threshold area
option.
6.1.3 Threshold setting automatically applied by SystemProvisioning linkage
1. Threshold setting automatically applied by VM Optimized Placement
When the configuration information for the group for which the above VM Optimization is enabled is
reflected on System Monitor - Performance Monitoring Services of SystemProvisioning, threshold
settings that are tailored to the VM Optimization settings and which are specified in
SystemProvisioning are automatically created and applied to all the machines in the group. The
automatically applied threshold setting can be checked in the Set Threshold Monitor dialog box.
The performance indicator of the performance monitor setting to be added
must be set to collect from the target node. Note
140
The automatically applied threshold settings by VM Optimized Placement are listed in blue in the Set
Threshold Monitor dialog box. In this dialog box, it is impossible to delete an automatically applied
threshold setting or to manually add a monitoring setting using the automatically added threshold setting,
but it is possible to change the notification setting for the threshold method.
141
2. Threshold setting automatically applied by Performance Monitor settings
When the configuration information for the group/model for which the Performance Monitor is enabled
in SystemProvisioning is reflected on System Monitor - Performance Monitoring Services, threshold
settings that are tailored to the monitoring profile settings and which are specified in
SystemProvisioning are automatically created and applied to all the machines in the group. The
automatically applied threshold setting can be checked in the Set Threshold Monitor dialog box.
142
The automatically applied threshold settings by Performance Monitor settings are listed in blue in the
Set Threshold Monitor dialog box. In this dialog box, it is impossible to delete an automatically applied
threshold setting or to manually add a monitoring setting using the automatically added threshold
setting. And, in System Monitor - Performance Monitoring Services, it is not allowed to change the
threshold value and the notification setting for the threshold method.
143
Chapter 7 Management of Performance Data
7.1 Data Management Tool
This section describes how to change the data storage period and perform data summarization.
1. On the Start menu, point to SigmaSystemCenter - Tools and then click System Monitor Data Management
Tool to start the System Monitor Data Management Tool.
2. To change the storage period for the summary data, select the summary interval to be changed in the list and
then click Modify to display the Summary Setting dialog box. Change the storage period, and then click OK.
144
3. To change the storage period for the collected data, change the collected data storage period setting for the
System Monitor Data Management Tool.
4. Click Run to change the data storage period and then perform data summarization.
Note
Make these settings so that the storage periods for the summary data items
satisfy the following conditions.
Storage period for 5 minutes ≤ Storage period for 15 minutes ≤ Storage period
for 1 hour ≤ Storage period for 1 day
Note
Before executing data summarization, stop System Monitor - Performance
Monitoring Services.
Note
Recompilation may take some time depending on the amount of data to be
recompiled. Keep this in mind with scheduling recompilation.
Note
If the warning value for the database size is exceeded during recompilation,
recompilation is stopped. At this time, reduce the data storage period and
then perform recompilation again.
145
5. In the confirmation dialog box that appears upon the completion of processing, click OK to close the System
Monitor Data Management Tool.
146
7.2 Database Size Monitoring
7.2.1 Database size warning setting
The database size monitoring function checks the database size every hour while the Performance
Monitoring Service is active. If the database size exceeds the specified value, a warning is issued. The default
database size at which a warning is issued is 3.6 GB. This alarm value can be changed. Change the alarm
value as described below.
1. Start up the management console for System Monitor - Performance Monitoring Services.
2. Right-click the management server name in the navigation tree window of the management console main
window to display the pop-up menu.
3. Click Set Configuration.
4. Click the Database Size tab.
5. Change the warning value.
6. Click OK to close the Configuration dialog box.
147
7.2.2 Database size warning
If the database size exceeds the specified warning value, the following message appears in the log window of
the connected management console, and is also recorded in the event log.
- Type: Warning
- Event ID: 67
- Message:
The database size has exceeded warning value WarningSizeGB. The current size is CurrentSizeGB.
Note
The database monitoring function monitors the size of the database area used
by System Monitor - Performance Monitoring Services. The size of other
database areas in the same database instance is not included in the target
database size.
Depending on SQL Server settings, the database size might not show the real
size immediately after data have been deleted from it. In this case, manually
compress the database file as needed. Note
148
Chapter 8 Command-Line Interface
8.1 ssc-perf Command
The ssc-perf command can display the performance data collected by System Monitor - Performance Monitoring
Services on the command prompt. This makes it possible to track the performance trends without starting the
management console. The ssc-perf command module is installed in the directory shown below:
<Installation Directory>\bin (Default value:%ProgramFiles%\NEC\SystemMonitorPerformance\bin)
An administrator privilege is needed to run the ssc-perf command.
8.1.1 The Basic Syntax
This section explains the basic syntax of the ssc-perf command. Help is displayed if parameters are omitted.
<Syntax>
<Parameters>
The following values can be specified for the option:
Option Description
-h Display help.
-v Display version information.
ssc-perf option | command subcommand [parameter] …
If the User Account Control, UAC, is valid, you need to execute with the
Administrator mode. For example, right-click Command Prompt and click
Run as administrator to launch the ssc-perf commands.
Note
149
8.1.2 show performancedata Command
With this command, the specified performance data from those registered to System Monitor - Performance
Monitoring Services is output in the CSV file format.
To output the performance data, you must specify the following items:
• Output period
• Output interval
• Node
• Performance indicator
• Statistical method
With these items specified, the performance data of the specified node, performance indicator and statistical
method at a specified output interval for a specified output period is output per line in the CSV file format. You
can specify so that the multiple performance data of Node/Performance indicator/Statistical method will be
output per column in one line. However, you can specify only one of the Node/Performance indicator/Statistical
methods.
If no data correspond to those you specified in the above, the performance data will not be output. If the data
partially include hours that the performance data are not obtained, the line with the data will be output but the
line without the data will not.
<Syntax>
<Parameters>
Parameter Required
/Optional
Description
-node Node
*2
Required
*1
Specify the path of the node on System Monitor - Performance
Monitoring Services whose performance data you want to
output.
Sample:
-node "GroupName\MachineName"
If you specify the path to the group without specifying the
machine name, the statistic information of the group is output.
-path Path
*2
Required
*1
Specify the path of the node on SigmaSystemCenter, whose
performance data you want to output in the following format:
Path of SystemProvisioning in the group setting of System
ssc-perf show performancedata {-node Node | -path Path} -indicator Indicator
-statistic Statistic [parameter]…
150
Parameter Required
/Optional
Description
Monitor - Performance Monitoring Services\Machine name
Sample:
-path
"CategoryName\GroupName\ModelName\MachineName"
If you do not use a model (machine in the default model), omit
the model name.
If you specify the path to the group or model without specifying
the machine name, all machines in the group or model are
specified.
-indicator
Indicator
*2
Required Specify the title of the performance indicator whose
performance data you want to output.
-statistic
Statistic
*2
Required Specify the title of statistical method whose performance data
you want to output.
The following values can be specified:
Value Description
Ave Displays the average value.
Max Displays the maximum value.
Min Displays the minimum value.
WeightAve Displays the weighted average value.
UpperRange Displays the average + standard deviation
value.
LowerRange Displays the average - standard deviation
value.
Sum Displays the summation value.
[-start StartTime] Optional Specify the start date and time of output period for the
performance data.
If you omit StartTime, the start date and time is determined
based on the EndTime value and the Period value as follows:
・ If omitted only StartTime, the start date and time is
determined by the following formula: EndTime-Period
・If omitted both StartTime and EndTime, the start date and
time is determined by subtracting Period from the time to
execute the command (see the description of EndTime).
For example, if the time to execute the command is "01:02", you
151
Parameter Required
/Optional
Description
specify "00:05" for Interval, and "01:00" for Period, StartTime
becomes "00:00."
This format depends on the date and time formats in the
regional and language options of Control Panel of the Windows
operating system on the management server.
Specify the date and time in the same format as that in your
operating system.
[-end EndTime] Optional Specify the end date and time of output period for the
performance data.
If you omit EndTime, the end date and time is determined as
follows:
・ If omitted only EndTime, the start date and time is
determined by the following formula: StartTime+Period
・If omitted both StartTime and EndTime, the end date and
time is the time rounded to the nearest Interval before the time
the command was executed as the following example shows.
For example, if the time to execute command is "01:02", you
specify "00:05" for Interval, and "01:00" for Period, EndTime
becomes "01:00."
[-period Period] Optional Specify the period from the start date and time to the end date
and time to output the performance data.
In the figure above, specify from 00:00 to 01:00 as Period.
・If you specified both StartTime and EndTime, the Period
value you specified is ignored and the actual period is
determined by the following formula: EndTime-StartTime
If omitted, the default is 01:00.
Interval
Period
00:00 00:05 00:10 00:50 00:55 01:00
Execute time is "01:02"
152
Parameter Required
/Optional
Description
Format: d | [d.]hh:mm[:ss]
d: The number of days. If you specify the value for "d", the
result of the formula d * 24 is added to the number of hours.
hh: The number of hours
mm: The number of minute
ss: The number of seconds
[-interval
Interval]
Optional Specify the interval of the data that are output per line.
In the figure above, the data are output every five minutes. So
specify the interval as five minutes.
This determines the performance data type on the database
that is used when outputting the data.
The performance data on the database include four types of
summary data and each type has different data interval.
The data with the largest interval in intervals smaller than the
Interval value you specified will be used. For details of the
summary data, see 1.4.2 "Usage of Summary Data."
If omitted, the default is 00:01.
Format: d | [d.]hh:mm[:ss]
For more information about the format to specify this option,
see the description about -period.
E.g.) The performance data that will be used according to the
-interval value you specified:
If you specify "00:04:59" for –interval, the collected data will be
used.
If you specify "00:05" for –interval, the 5-minute summary data
will be used.
If you specify "00:15" for –interval, the 15-minute summary
data will be used.
If you specify "01:00" for –interval, the 1-hour summary data
will be used.
If you specify "1" for –interval, the 1-hour summary data will
be used.
If you specify "2" for –interval, the 1-day summary data will be
used.
[-port PortNumber] Optional Specify the port number of the service to connect.
If omitted, the default is 26200.
153
Parameter Required
/Optional
Description
[-tab | -t] Optional Use a tab as a delimiter in displaying performance data. If
omitted, use a comma as a delimiter.
*1 Either -node or -path is required. These two parameters cannot be specified at the same
time.
*2 You can specify multiple values for only one of the three required parameters (-node (or
-path), -indicator and -statistic).
< Sample Output>
C:\Documents and Settings\Administrator>ssc-perf show performancedata -node
Group1\machine1 Group1\machine2 -indicator "CPU Usage (%)" -statistic Ave -period
00:10
Period 1/1/2010 00:00:00 AM-1/1/2010 00:10:00 00:10:00 AM Indicator:CPU Usage (%)
StatisticalMethod:Average
DateTime , Group1\machine1, Group1\machine2
1/1/2010 00:01:00 AM, 3.459734 , 3.534981
1/1/2010 00:02:00 AM, 7.41076 , 1.907642
1/1/2010 00:03:00 AM, 12.47074 , 0.7593492
1/1/2010 00:04:00 AM, 23.86721 , 3.245667
1/1/2010 00:05:00 AM, 27.60608 , 1.271073
1/1/2010 00:06:00 AM, 24.14691 , 7.729072
1/1/2010 00:07:00 AM, 23.14114 , 10.35331
1/1/2010 00:08:00 AM, 9.133735 , 8.398178
1/1/2010 00:09:00 AM, 4.893562 , 1.193273
1/1/2010 00:10:00 AM, 4.379085 , 0.8400003
C:\Documents and Settings\Administrator>
154
Chapter 9 Troubleshooting
9.1 The Event Log
System Monitor - Performance Monitoring Services records errors and operation event information in the event
log on the management server under the event log name "System Monitor - Performance Monitoring Services".
A list of event logs is shown below.
9.1.1 Event Logs for Performance Monitoring Service
source name: SystemMonitorPerformanceService ID Type category Description
1 Information Service Started Performance Monitoring Service. 2 Information Service Stopped Performance Monitoring Service. 3 Information Service Paused Performance Monitoring Service. 4 Information Service Restarted Performance Monitoring Service. 5 Error Service Error when starting Performance Monitoring Service. 6 Error Service Error when stopping Performance Monitoring Service. 7 Error Service Error when pausing Performance Monitoring Service. 8 Error Service Error when restarting Performance Monitoring Service. 9 Information Management
console
The user(account: domain\account) from the machine ‘machine’
logged out. 10 Error Data collecting The specified performance data (title: 'title', category name: ’category’,
instance name: ’instance’, counter name: ’counter’) cannot be collected
on the monitored machine ‘Machine’. This type of performance data is
not available for machine ‘Machine’. 11 Error Data collecting Failed to collect performance data (title: 'title', category
name: ’category’, instance name: ’instance’, counter name: ’counter’)
on the monitored machine ‘Machine’. 12 Error Other Failed to connect to the database (Server: Server\Instance,
DataBase: DataBase). 13 Error Other Failed to update the database. 14 Error Other Failed to initialize the data. 15 Error Other Failed to reference the database. 16 Warning Service Failed to save the Performance Monitoring Service settings. Will read
the previous settings at the next activation.
155
ID Type category Description
17 Warning Service Failed to read the Performance Monitoring Service settings. Will read
the initial settings. 18 Error Management
console
Failed to logon to the account (domain\account) from the machine
‘machine’. 19 Warning Data collecting Cannot access the monitored machine ‘Machine’. 20 Error Other Failed to initialize the logging facility. 21 Error Other Failed to stop the logging facility. 22 Information Management
console
The user (account: domain\account) logged on from the machine
‘machine’. 23 Error Management
console
An unknown user account attempted to connect.
24 Error Service Error in termination processing. 25 Error Service Error processing failed. 26 Error Other Internal error: message 27 Error Management
console
Cannot connect to the Management Console (Uri: URL).
28 Error Other Failed to delete performance data after the storage period. 29 Error Service Stopped Performance Monitoring Service because an irrecoverable
error was detected. 30 Error SystemProvisioning
relation
Cannot use SystemProvisioning.
31 Error SystemProvisioning
relation
Error when applying SystemProvisioning Configuration.
32 Error Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (System provisioning
group path: ’PvmGroupPath’) exceeded the upper critical threshold
value or fell below the lower critical threshold value (Value). 33 Error Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the
machine ‘Machine’ (SystemProvisioning path: ’PvmPath’) exceeded
the upper critical threshold value or fell below the lower critical
threshold value (Value). 34 Warning Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistic’s) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) exceeded the upper warning threshold
value or fell below the lower warning threshold value (Value).
156
ID Type category Description
35 Warning Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the
machine ‘Machine’ (SystemProvisioning path: ’PvmPath’) exceeded
the upper warning threshold value or fell below the lower warning
threshold value (Value). 36 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) recovered from the critical value
(Value). 37 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the machine
‘Machine’ (SystemProvisioning path: ’PvmPath’) recovered from the
critical value (Value). 38 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) recovered from the warning value
(Value). 39 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the machine
‘Machine’ (SystemProvisioning path: ’PvmPath’) recovered from the
warning value (Value). 40 Error Performance
information
No performance indicator (title: ’Title’) specified for the OS of the
object machine (machine name: ’Machine’). 41 Warning Data collecting Failed to collect performance data (title: ‘Title’, resource: ’Resource’,
instance: ’Instance’, and performance index: ‘PerformanceIndex’)
about the monitored machine ‘Machine’. 42 Error SystemProvisioning
relation
Failed to notify SystemProvisioning.
43 Error Performance
information
Illegal specification of the performance indicator (title: ’Title’)
44 Information SystemProvisioning
relation
Applied the SystemProvisioning Configuration.
45 Information Data collecting Already connected to the target machine 'Machine' (shared resource:
'Resource'). Use this connection for collecting performance data. 46 Error Data collecting Failed to disconnect to the target machine 'Machine' (shared resource:
'Resource'). 47 Information System
configuration
Added machine 'Machine' to group 'Group'.
48 Information System
configuration
Removed machine 'Machine' from group 'Group'.
157
ID Type category Description
49 Information System
configuration
Moved machine 'Machine' from group 'Group' to group 'Group'.
50 Information System
configuration
Changed setting for machine 'Machine'.
51 Error Data collecting Machine 'Machine' had an access error. 52 Information Data collecting Machine 'Machine' recovered from an access error. 53 Information System
configuration
Added group 'Group'.
54 Information System
configuration
Removed group 'Group'.
55 Information System
configuration
Changed setting for group 'Group'.
56 Information Setting Changed settings for collecting performance data. 57 Information Setting Changed settings for threshold monitoring. 58 Information Setting Changed settings for configuration. 62 Warning Data collecting No performance data (title: 'Title') about machine 'Machine' existed.
Calculated performance data about group 'Group' at 'Time' without
performance data about machine 'Machine'. 63 Warning Data collecting Cannot finish collecting performance data (title: 'Title') about
machine 'Machine' in time (Time-Time) 64 Error Other Failed to output a debug log. 65 Warning SystemProvisioning
relation
Made a notification to SystemProvisioning on the node 'Node'.
SystemProvisioning Configuration Application is not enabled for the
node 'Node'. 66 Error Data collecting An error occurred in the connection to the target machine 'Machine'. 67 Warning Other The database size has exceeded warning value WarningSizeGB. The
current size is CurrentSizeGB. 69 Information Data collecting The collecting of the performance data (title: 'Title') is restarted on
the monitored machine 'Machine'. 70 Error Data collecting Failed to calculate performance data (machine: 'Machine', title: 'Title',
resource: 'Resource', instance: 'Instance', performance index: 'Index'). 71 Information Data collecting Paused collecting performance data of machine 'Machine'. 72 Information Data collecting Restarted collecting performance data of machine 'Machine'. 73 Information Data collecting Paused collecting performance data of group 'Group'. 74 Information Data collecting Restarted collecting performance data of group 'Group'. 75 Warning Data collecting Failed to calculate performance data (machine: 'Machine', title: 'Title',
Category: 'Category', Instance: 'Instance', Counter: 'Counter').
158
ID Type category Description
76 Information Data collecting Data provider (server:Server, type:Type) recovered from an access
error. 77 Error Data collecting Data provider (server:Server, type:Type) had an access error. 78 Error Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (System provisioning
group path: ’PvmGroupPath’) exceeded the upper critical threshold
value or fell below the lower critical threshold value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 79 Error Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the
machine ‘Machine’ (SystemProvisioning path: ’PvmPath’) exceeded
the upper critical threshold value or fell below the lower critical
threshold value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 80 Warning Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistic’s) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) exceeded the upper warning threshold
value or fell below the lower warning threshold value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 81 Warning Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the
machine ‘Machine’ (SystemProvisioning path: ’PvmPath’) exceeded
the upper warning threshold value or fell below the lower warning
threshold value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 82 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) recovered from the critical value
(Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 83 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the machine
‘Machine’ (SystemProvisioning path: ’PvmPath’) recovered from the
critical value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning.
159
ID Type category Description
84 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’
(Statistics: ’Statistics’) of the group ‘Group’ (SystemProvisioning
group path: ’PvmGroupPath’) recovered from the warning value
(Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 85 Information Threshold
monitoring
The performance indicator ‘PerformanceIndicatorTitle’ of the machine
‘Machine’ (SystemProvisioning path: ’PvmPath’) recovered from the
warning value (Value).
Reported [Event Category]eventcategory [Event]event to
SystemProvisioning. 86 Error Data collecting Failed to collect performance data (title: 'title', executing string:
'executingstring') about the monitored machine 'machine'. 90 Error Data collecting The specified performance data (title: 'title', metric group:
'metricgroup', object: 'object', counter: 'counter') cannot be collected on
the monitored machine ''machine''. This type of performance data is
not available for machine ''machine''. 91 Error Data collecting The specified performance data (title: ''title'', resource: 'resource',
instance: 'instance', performance index: 'performanceindex') cannot be
collected on the monitored machine 'machine'. This type of
performance data is not available for machine ''machine''.
9.1.2 Event logs for Management Console
source name: SystemMonitorPerformanceConsole
ID Type category Description
75 Information Management
console
Connected to Management Server 'Server'.
76 Information Management
console
Disconnected to Management Server 'Server'.
77 Error Management
console
An error occurred in the connecting/disconnecting to the Management
Server 'Server'.
84 Error Management
console
Internal error: message
85 Information Management
console
Started management console.
86 Information Management
console
The Management Console is closed.
160
ID Type category Description
89 Error Management
console
Failed to initialize the connection. Change the settings of the Management
Console.
90 Error Management
console
The time difference between the management server Server and this
console machine is larger than Time minutes.
By default, previous logs are overwritten when the maximum log size is
reached. To change these default settings, configure the Event Viewer
settings. See "9.10. 3. Situations where logs are not recorded in the event
log" for information on how to change the Event Viewer settings.
Note
161
9.2 What to do if the Performance Monitoring Service does not Start
After System Monitor - Performance Monitoring Services is installed, the Performance Monitoring Service
(System Monitor Performance Monitoring Service) should start when the management server is restarted.
If the Performance Monitoring Service does not start, this could be due to one of the causes listed as (1) to (7)
below. Isolate the cause by checking the message that is output when the Performance Monitoring Service fails,
and then take the action listed below. Error messages are also displayed when the Performance Monitoring
Service is started manually. Refer to (7) if you enabled the System cryptography: Use FIPS compliant
algorithms for encryption, hashing, and signing policy in Local Computer Policy on the management server.
Refer to (3), (4), (5), and (6) if you changed the execution account from the local system account.
When execution account is changed from a local system account, confirm the following.
• "User must change password at next logon" option has not been added.
• The password has not expired.
(1) The Database Settings are not correct
<Messages that are output>
• Event log Log Type Source ID Description Notes
Application Error SystemMonitor Performance Service
0 Service cannot be started.
System.Data.SqlClient.SqlException: Cannot open database requested in login 'RM_PerformanceDataBase2'. Login fails.
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
5 Error when starting Performance Monitoring Service. Details: Cannot open database requested in login 'RM_PerformanceDataBase2'. Login fails.
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
12 Failed to connect to the database (Server: Server\Instance , DataBase:DataBase:). Details: Cannot open database requested in login 'RM_PerformanceDataBase2'. Login fails.
• Error messages Message Notes
Could not start the SystemMonitor Performance Service service on Local Computer. The service did not return an error. This could be an internal Windows error or an internal service error. If the problem persists, contact your system administrator.
Windows 2000 only
The System Monitor Performance Monitoring Service service on Local Computer started and then stopped. Some services stop automatically if they have no work to do, for example, the Performance Logs and Alerts service.
Windows Server 2003 only
162
<Action>
The database may not have been set up correctly, or a database file may be damaged. Reinstall System
Monitor - Performance Monitoring Services.
(2) The Database cannot be Referenced or Updated
See "9.9.3 Errors and Actions for Database Faults".
(3) The Execution Account and Password Settings for the Performance Monitoring Service are not correct
<Messages that are output>
• Event log Log Type Source ID Description Notes
System Error Service Control Manager
7000 The System Monitor Performance Monitoring Service service failed to start due to the following error: The service did not start due to a logon failure.
System Error Service Control Manager
7013 Logon attempt with current password failed with the following error: Logon Failure: Unknown Username or bad password
Windows 2000 only
System Error Service Control Manager
7038 The SystemMonitor Performance Service service was unable to log on as domain\account with the currently configured password due to the following error: Logon Failure: unknown user name or bad password.
Windows Server 2003 only
• Error messages Message Notes
Could not start the System Monitor Performance Monitoring Service service on Local Computer. Error 1069: The service did not start due to a logon failure.
<Action>
Set the execution account and password for the Performance Monitoring Service correctly. See "2.5 The
Execution Account for the Performance Monitoring Service" for information about execution account.
163
(4) The Execution Account for the Performance Monitoring Service does not have OS Administrator Authority
<Messages that are output>
• Event log Log Type Source ID Description Notes
Application Error SystemMonitor Performance Service
0 The service did not start. System.Security.SecurityException: Requested registry access is not allowed.
• Error messages Message Notes
Could not start the SystemMonitor Performance Service service on Local Computer. The service did not return an error. This could be an internal Windows error or an internal service error. If the problem persists, contact your system administrator.
Windows 2000 only
The System Monitor Performance Monitoring Service service on Local Computer started and then stopped. Some service stop automatically if they have no work to do, for example, the Performance Logs and Alerts service.
Windows Server 2003 only
<Action>
The execution account for the Performance Monitoring Service must have OS administrator authority for
the management server. Specify an account that has OS administrator authority as the execution account
for the Performance Monitoring Service. See "2.5 The Execution Account for the Performance Monitoring
Service" for information about execution account.
(5) The Execution Account for the Performance Monitoring Service does not have "Log on as a Service" Rights
<Messages that are output>
• Event log Log Type Source ID Description Notes
System Error Service Control Manager
7000 The System Monitor Performance Monitoring Service service failed to start due to the following error: The service did not start due to a logon failure.
System Error Service Control Manager
7013 Logon attempt with current password failed with the following error: Logon failure: the user has not been granted the requested logon type at this computer.
Windows 2000 only
System Error Service Control Manager
7038 The SystemMonitor Performance Service service was unable to log on as domain\account with the currently configured password due to the following error: Logon failure: the user has not been granted the requested logon type at this computer.
Windows 2003 only
164
• Error messages Message Notes
Could not start the System Monitor Performance Monitoring Service service on Local Computer. Error 1069: The service did not start due to a logon failure.
<Action>
The execution account for the Performance Monitoring Service must have "Log on as a service" rights.
Add this right, referring to "2.5.3 Rights for the Performance Monitoring Service Execution Account”.
Also, if the account has "Deny log on as a service" rights, delete these rights.
(6) An Account with an Empty Password has been specified as the Execution Account for the Performance
Monitoring Service
This error only occurs if the OS is Windows Server 2003.
<Messages that are output>
• Event log Log Type Source ID Description Notes
System Error Service Control Manager
7000 The System Monitor Performance Monitoring Service service failed to start due to the following error: The service did not start due to a logon failure.
Windows Server 2003 only
System Error Service Control Manager
7038 The SystemMonitor Performance Service service was unable to log on as domain\account with the currently configured password due to the following error: Logon failure: user account restriction. Possible reasons are blank passwords not allowed, logon hour restrictions, or a policy restriction has been enforced.
Windows Server 2003 only
• Error messages Message Notes
Could not start the System Monitor Performance Monitoring Service on Local Computer. Error 1069: The service did not start due to a logon failure.
Windows Server 2003 only
<Action>
With Windows Server 2003, it is possible to specify the security option "Limit local account use of blank
passwords to console logon only". If this option is enabled, accounts that have blank passwords cannot be
used as the execution account for the Performance Monitoring Service. Use an account for which a
password has been set up.
165
(7) The policy "System cryptography: Use FIPS compliant algorithms for encryption, hashing, and signing",
which is in Security Settings in Local Computer Policy, is enabled on the management server
<Messages that are output>
• Event log Log Type Source ID Description Notes
Application Error SystemMonitor Performance Service
0 Service cannot be started. System.InvalidOperationException: This implementation is not part of the Windows Platform FIPS validated cryptographic algorithms.
• Error messages Message Notes
The System Monitor Performance Service service on Local Computer started and then stopped. Some services stop automatically if they are not in use by other services or programs.
<Action>
RC2 is the default encryption algorithm used in SystemMonitor Performance Service. The algorithm is not
compliant with FIPS 140-2 cryptographic algorithm, and, because of this, it causes an error when a service
is started. You can work around this issue as follows:
1. On Local Computer Policy, disable the security setting of System cryptography: Use FIPS compliant algorithm for
encryption, hashing, and signing, and then restart the System Monitor Performance Monitoring Service.
2. If your business security policy does not allow for you to disable System cryptography: Use FIPS compliant
algorithm for encryption, hashing, and signing, perform the following procedure to change the encryption
algorithm:
(1) Disable the System cryptography: Use FIPS compliant algorithm for encryption, hashing, and
signing setting, and then return System Monitor Performance Monitoring Service to the state
in which it can use the original encryption algorithm.
(2) On the System Monitor Performance Monitoring Service management server, add the
following appSettings setting immediately under the configuration tag of .Net Framework's
machine config (%windir%\Microsoft.NET\Framework64\v4.0.30319\Config) to change the
encryption algorithm used in System Monitor Performance Monitoring Service from RC2 to
3DES. After changing the setting, start the System Monitor Performance Monitoring Service.
<appSettings>
<add key="SystemMonitorPerformanceMonitoringEncryption"
value="3DES"/>
</appSettings>
166
(3) On the SigmaSystemCenter management server, add the following appSettings setting
immediately under the configuration tag of .Net Framework's machine config
(%windir%\Microsoft.NET\Framework\v4.0.30319\Config) to change the encryption
algorithm used to connect to System Monitor Performance Monitoring Service from RC2 to
3DES. If the file is same as that changed in (2), the configuration change is not needed. After
changing the setting, start the PVMService.
<appSettings>
<add key=" SystemMonitorPerformanceMonitoringEncryption"
value="3DES"/>
</appSettings>
(4) Enable the System cryptography: Use FIPS compliant algorithm for encryption, hashing, and
signing setting.
167
9.3 What to do if the Connection from Management Console to Server Fails
The following lists error messages output when connection from the management console to the management
server failed, and actions to be taken. Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceConsole
77 An error occurred in the connecting/disconnecting to the Management Server 'Server'. Details: No connection could be made because the target machine actively refused it.
Action
• Check whether the management server is running. • Start the Performance Monitoring Service on the management server if it is not running already. • Check whether there are any network problems between the management server and the management
console machine. • Check whether the port number for the Performance Monitoring Service has been changed. If so, specify the
new port number.
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceConsole
77 An error occurred in the connecting/disconnecting to the Management Server 'Server'. Details: Logon failure.
Action
• Check the account and password specifications. • Accounts that do not have OS administrator authority cannot log on. Check whether the account has OS
administrator authority. • If either the logon account or the groups to which it belongs has been set to any of the following in Security
Settings, the account cannot logon. Check the Security Settings of the logon account and the groups it belongs to. ① Neither the user account nor any of the groups to which it belongs has been granted the Access this
computer from the network user right. ② The user account or one or more of the groups it is a member of has been listed in the Deny access to
this computer from the network. • Only if the OS for the management server is Windows 2000.
If the execution account for the Performance Monitoring Service does not have the "Act as part of the operating system" right, add this right, referring to" 2.5.3 Rights for the Performance Monitoring Service Execution Account". After the Performance Monitoring Service has stopped, it is possible to log on.
• With Windows Server 2003, it is possible to specify the security option "Limit local account use of blank passwords to console logon only". If this option is enabled, accounts that have blank passwords cannot be used as the log on account. Use an account for which a password has been set up.
168
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceConsole
89 Failed to initialize the connection. Change the settings of the Management Console. Details: Only one usage of each socket address (protocol/network address/port) is normally permitted.
Action
• The port number used for communications from the Performance Monitoring Service to the management console is already being used by another product. Change the port number in management console information.
• There is a possibility that multiple management consoles are running on a console machine. If multiple management consoles are already running, verify that each management console has different port number settings for its management console information.
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceConsole
77 An error occurred in the connecting/disconnecting to the Management Server 'Server'. Details: Illegal setting on management console information. Connection timeout due to no response from Performance Monitoring Service.
Action
• Check the host name setting for the management console information. Verify that the management server can access the management console machine using the specified host name.
• Check whether there are any problems with the network between the management server and the management console machine.
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceConsole
77 An error occurred in the connecting/disconnecting to the Management Server 'Server'. Details: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
Action
• Check whether there are any problems with the network between the management server and the management console machine.
169
9.4 What to do if Performance Data Collecting Fails
If the collecting of performance data fails, refer to the following actions. The collecting status for performance
data is displayed using icons in the navigation tree window. See "4.4 Checking the Collecting Status for
Performance Data" for the meanings of these icons.
The navigation tree window:
Collecting failed for some Performance Indicator items (performance indicator error status)
The machine cannot be accessed or collecting failed for all Performance Indicators for the machine
(access error status)
In Windows machines, VMware ESX/ESXi virtual machine servers or virtual machines data collecting, when
Category and Counter is specified correctly but Instance is incorrect, the performance data will not be collected.
But the monitored machine will not be set to be an error status. If performance data is not being collected even
though no errors are recorded in the event log and the icon display does not indicate an error, then verify that
the instance name for the Performance Indicator is correct.
170
9.4.1 Errors and Actions for when Performance Data Collecting Fails
The event logs for when performance data collecting fails and the actions to take are listed below. Log Type Source ID Description
System Monitor - Performance Monitoring Services
Warning SystemMonitor PerformanceService
19 Cannot access the monitored machine 'Machine'.
System Monitor - Performance Monitoring Services
Warning SystemMonitor PerformanceService
41 Failed to collect performance data (title: 'title', resource: 'Resource', instance: 'Instance', and performance index: 'PerformanceIndex') about the monitored machine 'Machine'.
Action
• Check the specifications for the machine name or IP address. • The Performance Monitoring Service accesses monitored machines using the specified account name and
password. Check the following points concerning the user accounts for the monitored machine: • An account with the same account name and password as the execution account has been registered on the
monitored machine. • The password for the account has not expired.
• Check the following when the SystemProvisioning configuration is applied to System Monitor - Performance
Monitoring Services: • System Monitor - Performance Monitoring Services uses the Management IP address already set up using
SystemProvisioning to connect to the target machines. Specify the Management IP address in the Machine Settings window with the SystemProvisioning Web Console. However, for some system configurations, there are cases where it is not possible to connect to monitored machines using the Management IP addresses acquired, and so if a monitored machine produces an error, disable [Apply IP address information] option in the Group Settings window and set the IP address using the Machine Settings window of the System Monitor - Performance Monitoring Services management console. When the IP address is not specified, System Monitor - Performance Monitoring Services uses the machine name to connect to the target machines. If the monitored machine is a Linux machine, VMware ESX/ESXi or Citrix XenServer, the machine name must be registered with either DNS or the “hosts” file on the System Monitor - Performance Monitoring Services management server.
• Check the OS name on the machine settings. When this OS name is incorrect, check the configuration settings on SystemProvisioning.
• If the monitored machine is running a Windows operating system:
The Performance Monitoring Service accesses monitored machines using the specified account name and password. According to the settings, the Performance Monitoring Service accesses monitored machines using the service execution account. For conditions to be used the service execution account as an access account, see "2.5.1 Setting up the Performance Monitoring Service Execution Account". Check the following points concerning the user accounts for the monitored machine: • This account belongs to the Administrator group or Performance Monitor Users group of the monitored
machine. • The "User must change password at next logon" option has not been added to the account. • This account does not have a blank password.
With Windows Server 2003, it is possible to specify the security option "Limit local account use of blank passwords to console logon only". If this option is enabled, accounts that have blank passwords cannot be used. Use an account for which a password has been set up.
• If either the account or the groups to which it belongs has been set to any of the following in Security Settings, the account cannot access the monitored machine. Check Security Settings of the account and the groups it belongs to. - The "Access this computer from the network" right is not assigned to both the account and the group
which the account belongs to.
171
- The "Deny access to this computer from the network" right list contains the account or one or more groups which the account belongs to.
Start the following service for monitored machines. • Remote Registry • Server
Check the following points concerning the connection to monitored machines: • If the Windows firewall function for the monitored machine has been enabled, make exception settings for
the Windows firewall by referring to "1.7.4 Ports Used between Management Servers and Monitored Machines".
If the monitored machine is running a Windows XP, change local security policy settings as follows:
1. Start the Local Security Policy using the Administrative Tools in Control Panel
2. The Local Security Settings are appeared. Select Security Options of Local Policies from the tree on the
left side.
3. Display the properties by double-clicking Network access: Sharing and security model for local accounts.
4. Select Classic – local users authenticate as themselves and then Click OK. • If the monitored machine is running a Linux OS/KVM: • SSH service has started. • Authentication protocol is set correctly. • SSH communications have been exempted from firewall blocking.
Check that an IP address has been set up on the monitored machine settings, or the name of the monitored machine has been registered with DNS or the "hosts" file on the management server. • If VMware ESX/ESXi is the monitored machine (VMware Web Service is used to connect):
Check the following points concerning the user accounts for the monitored machine: • An account with the same account name and password as the specified account has been registered on the
monitored machine. • A role other than "No Access" is assigned to the connection account on the VMware ESX/ESXi.
Check the following points: • An IP address has been set up on the monitored machine settings. • VMware Web Service has started on the VMware ESX/ESXi. • SSL communications have been exempted from firewall blocking.
• If the monitored machine is a virtual machine running on a VMware ESX/ESXi. Check the following points: • The virtual machine and the VMware ESX/ESXi are monitored by the same management server. • Collecting performance data from the VMware ESX/ESXi is successfully performed.
• If Citrix XenServer is the monitored machine:
Check the following points: • An IP address has been set up on the monitored machine settings. • SSL communications have been exempted from firewall blocking. • The monitored machine is a pool master machine or a target of the SystemProvisioning configuration
application.
172
• Check whether the monitored machine is running. • Check whether there are any network problems between the management server and the monitored machine. • It may not be possible to access the monitored machine if it is under high load. • Check whether the monitored machine's OS is supported by System Monitor – Performance Monitoring
Services.
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
10 The specified performance data (title:'title', category name: 'category', instance name: 'instance', counter name: 'counter') cannot be collected on the monitored machine 'Machine'. This type of performance data is not available for machine 'Machine'. Details: *************************************
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
43 Illegal specification of performance indicator (title: 'title')
Action
Check the detailed information in the event log description (the section after "Details :"). • Fix any mistakes with the specification for Performance Indicators. • If "Details: Category does not exist" appears, possible causes are follows:
By default, data collecting on the monitored machine with Windows 2000 might fails, In this case, change the default execution account for the Performance Monitoring service. For how to do it, see 2.5, "The Execution Account for the Performance Monitoring Service."
• By default, the monitored machine with Windows 2000 cannot collect the data within the LocalDisk category. If you want to collect the data, run the following command to change the setting.
diskperf –y
After running the command, restart the monitored machine. Then, stop the Performance Monitoring service and restart it.
• The monitored machine with Windows 2000 cannot obtain the following performance data within the Processor category: % Idle Time, %C1 Time, %C2 Time, %C3 Time, C1 Transitions/sec, C2 Transitions/sec, C3 Transitions/sec
• If a Performance Indicator category is specified that is not supported by the operating system and applications on the monitored machine, it may not be possible to collect the performance data because the specified category does not exist on the monitored machine. The Performance Indicators that can be collected on a monitored machine depend on the type of OS and on the applications that are installed. Check whether the specified Performance Indicator can be collected. See "1.3 Collected Performance Data" for information on the categories that System Monitor - Performance Monitoring Services supports.
• If both System Monitor - Performance Monitoring Services and other performance monitoring products monitor one Windows monitored machine at the same time, the "Category does not exist" error might occur while collecting the performance data.
System Monitor - Performance Monitoring Services uses the Windows performance monitor feature to obtain the performance data from the Windows monitored machine. This issue occurs other monitoring products also use the Windows performance monitor feature as well as System Monitor - Performance Monitoring Services. Use either System Monitor - Performance Monitoring Services or other products to monitor the Windows monitored machine.
If the icon is displayed for a monitored machine, the machine is in a performance indicator error status. Change the Performance Indicator settings for the machine.
173
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
86 Failed to collect performance data (title: 'title',
executing string: 'executingstring') about the
monitored machine 'machine'. Details: *************************************
Action
Check the detailed information in the event log description (the section after "Details :"). Possible causes for
each problem are as follows. • Cannot access the monitored machine ‘Machine’. • Check the power status and the OS status of the machine. • Check the machine name or the IP address, and what OS name is specified. • Check that the machine is established as an SSH server. • Check that the SSH service is enabled. • When the firewall is enabled, check the firewall exception setting is configured.
(If not, configure the firewall exception setting seeing "1.7.4, Ports Used between Management Servers
and Monitored Machines".) • User authentication failed. • Check that the same account name and password with the specified account are registered. • Check that the appropriate authentication protocol is specified. • Check that the specified account has the SSH access right.
• Connection timeout. • Check what login shell is specified for the specified account. If the "cmd.exe" is not specified, specify it as the login shell.
• ’"@metric value" was not outputted from the script.’ or ‘"@metric value" was not outputted from the script for performance indicator {1} of the machine {0}.’ • If the performance indicator contains wrong execution strings specified, correct them. • The script is not written to output ”@metric value” to the standard output. Review the contents of the script seeing "1.3.3,
Custom Performance Indicator"
174
9.4.2 Error Status Recovery Processing
System Monitor - Performance Monitoring Services stops data collecting processing for monitored machines
that it has failed to access or for which all Performance Indicators failed to be collected. Machines where data
collecting processing has stopped are displayed using the icon. Machines where data collecting processing
has partially failed are displayed using the icon.
Remove the error cause to restart data collecting. To restart data collecting in arbitrary timing, perform error
status recovery. The procedure for error status recovery processing is as follows:
1. Right-click the machine name and click Change Data Collecting Status.
A window appears.
2. Click Recover from error status. When data collecting starts, the icon display for the machine will change to .
When error status recovery processing is performed for a group, error status recovery processing is executed for
every machine in the group.
The procedure for error status recovery processing for groups is as follows:
1. Right-click the group name and select Change Data Collecting Status. A window appears.
2. Click Recover from error status. When data collecting starts, the icon display for the group will change to and the
icon display for machines in the group will change to .
175
9.5 What to do if Performance Data Collecting is Delayed
If the collecting of performance data is delayed, refer to the following actions. What we call "Data Collecting
Delay" is the situation that a performance data collecting could not be finished within the specified period.
When this delay occurs, a discontinued graph line may appear on the graph window and a notification of
monitoring threshold may be delayed.
A warning log on the event log can be seen when the performance data collecting is delayed.
9.5.1 Errors and Actions for when Performance Data Collecting is Delayed
The event logs for when performance data collecting is delayed and the actions to take are listed below. Log Type Source ID Description
System Monitor - Performance Monitoring Services
Warning SystemMonitor PerformanceService
63 Cannot finish collecting performance data (title: 'Title') about machine 'Machine' in time (StartTime-EndTime).
Action
• The number of data to be collected within a specified period may be too much. Consider the following actions. - Increasing the data collecting interval - Reducing the number of Performance Indicator items that are collected - Reducing the number of machines that are monitored
• The load of the management server may be too high. Check the reason of this high load and exclude this
reason from your system. • An internal operation of Windows may let the data collecting operation take too long time when the OS of
the monitored machine is Windows Vista or Windows Server 2008 R1. If this entry is logged, consider to increase the data collecting interval.
176
9.6 What to do if there are SystemProvisioning Configuration Applying Errors
SystemProvisioning configuration information can be applied to System Monitor - Performance Monitoring
Services. If an error occurs during configuration applying process, identify the cause by checking the message that
has been output, and then take the appropriate action.
(1) Cannot access the SystemProvisioning management server
Log Type Source ID Description System Monitor -
Performance Monitoring
Services
Error SystemMonitorPerformanceService
30 Cannot use SystemProvisioning. Details: Cannot connect SystemProvisioning
Action
• Verify that the name of the SystemProvisioning management server is correct. Refer to Section, "2.6 SystemProvisioning Connection Settings" for information on how to make this setting.
• Verify that the SystemProvisioning management server is running correctly. • Verify that each SystemProvisioning service is running correctly. • If the Performance Monitoring Service and SystemProvisioning are not on the same management server and
a Windows Firewall has been enabled on the SystemProvisioning management server, open the port for SystemProvisioning UniversalConnector. Refer to Section, "2.6 SystemProvisioning Connection Settings" for more information on firewall settings.
• If the Performance Monitoring Service and SystemProvisioning are not on the same management server, verify that the Performance Monitoring Service execution account is same as the OS administrator account of SystemProvisioning management server.
• When SystemProvisioning is clustered with EXPRESSCLUSTER, the virtual computer name cannot be used as SystemProvisioning Management Server name.
(2) SystemProvisioning is not running correctly
Log Type Source ID Description System Monitor -
Performance Monitoring
Services
Error SystemMonitorPerformanceService
31 Error when applying SystemProvisioning Configuration. Details: Input data contains invalid character.
Action
• Verify that the version of SystemProvisioning being linked is supported by System Monitor - Performance Monitoring Services.
(3) Performance indicator does not exist
Log Type Source ID Description System Monitor -
Performance Monitoring
Services
Error SystemMonitorPerformanceService
31 Error when applying SystemProvisioning Configuration. Details: Specified performance indicator is not found.
Action
• The monitoring profile is specified on SystemProvisioning, but a performance indicator which does not exist on System Monitor - Performance Monitoring Service is included in it. • Verify whether the monitoring profile specified on SystemProvisioning is correct. • Verify whether the performance indicator is defined correctly on System Monitor - Performance Monitoring
Service.
177
9.7 What to do if there are Errors when Performance Abnormalities are Reported to SystemProvisioning
System Monitor - Performance Monitoring Services can use monitoring thresholds for Performance Indicators
that are collected to detect abnormalities in the load status of monitored machines and can report such
abnormalities to SystemProvisioning.
If an error occurs when an abnormality is reported, identify the cause by checking the message that has been
output, and then take the appropriate action.
(1) Cannot Access the SystemProvisioning Management Server
Log Type Source ID Description System Monitor -
Performance Monitoring
Services
Error SystemMonitorPerformanceService
42 Failed to notify SystemProvisioning. Details: System error: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
Action
• Verify that the name of the SystemProvisioning management server is correct. Refer to Section, "2.6 SystemProvisioning Connection Settings" for more information.
• Verify that the SystemProvisioning management server is running correctly. • Verify that the "System Monitor Service" has started on the SystemProvisioning management server. • If the Performance Monitoring Service and SystemProvisioning are not on the same management server and
a Windows Firewall has been enabled on the SystemProvisioning management server, open the port for System Monitor – Event Monitoring Services. Refer to Section, "2.6 SystemProvisioning Connection Settings" for more information on firewall settings.
• If the Performance Monitoring Service and SystemProvisioning are not on the same management server, verify that the Performance Monitoring Service execution account is same as the OS administrator account of SystemProvisioning management server.
• When SystemProvisioning is clustered with EXPRESSCLUSTER, the virtual computer name cannot be used as SystemProvisioning Management Server name.
(2) SystemProvisioning is not running correctly
Log Type Source ID Description System Monitor -
Performance Monitoring
Services
Error SystemMonitorPerformanceService
42 Failed to notify SystemProvisioning. Details: Could not resolve a call to the correct method.
Action
• Verify that the SystemProvisioning management server is running correctly. • Verify that each service on the SystemProvisioning management server is running correctly.
178
9.8 What to do if the SystemMonitor data management tool might not start
If an error message appears when you start the SystemMonitor data management tool, the tool might fail to
connect to the SQL Server database.
The following describes about possible causes and potential solutions for the error. Determine the cause based
on the circumstances in which the error occurred and resolve the error.
1. Cannot connect to the database
This error message appears if the database service that uses System Monitor - Performance Monitoring
Services is not started yet. If this message appears, start the SQL Server service as the following
description.
・ The database service starts automatically at startup of the Performance Monitoring Service. When
starting it manually, start the service whose name is "MSSQL$Instance Name" (the default instance
name is "SSCCMDB").
After checking the above and eliminating the problem, restart the data management tool.
2. The execution account does not have the privilege to connect to the database
If the execution account without the privilege to connect to the database attempts to connect to the SQL
Server database from the System Monitor data management tool, the following error occurs.
179
The user account to run the System Monitor data management tool must have the sysadmin privilege for
the database. Typically this issue occurs if the user account to run the System Monitor data management
tool is not same as that specified during SQL Server installation.
To resolve this issue, log on to the management server with the user account that has the sysadmin
privilege for the System Monitor - Performance Monitoring Services database, and grant the sysadmin
privilege to the user account to run the System Monitor data management tool by using the following
command.
>sqlcmd -E -S (local)\SSCCMDB
1> exec master..sp_addsrvrolemember @loginame = N'domain\account', @rolename =
N'sysadmin'
2> go
After checking the above and eliminating the problem, restart the data management tool.
180
9.9 Database
9.9.1 Estimating the Disk Space Required for Saving Performance Data
When 1 machine are to be monitored, there are 1 performance indicator, the collection interval is 1 minute,
and the default storage period is to be used, then the data size can be estimated as follows.
Collected data: 3 (days) * 60 KB = 180 KB
5-minute summary data: 7 (days) * 30 KB = 210 KB
15-minute summary data: 30 (days) * 7 KB = 210 KB
1-hour summary data: 3 * 30 (days) * 2 KB = 180 KB
1-day summary data: 5 * 365 (days) * 0.1KB = 182.5 KB
Total size: (180 +210 + 210 + 180 + 182.5) (KB) * 1 (machines)
* 1 (performance indicator)
= 962.5 (KB) = approximately 1 (MB)
9.9.2 Errors and Actions for when There is not Enough Disk Space
The event logs for when there is not enough disk space and the actions to take are listed below. Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
13 Failed to update the database. Details: Could not allocate space for object 'RM_PerformanceData' in database 'RM_PerformanceDataBase2' because the 'PRIMARY' filegroup is full.
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
29 Stopped Performance Monitoring Service because an irrecoverable error was detected. Details: Could not allocate space for object 'RM_PerformanceData' in database 'RM_PerformanceDataBase2' because the 'PRIMARY' filegroup is full.
Action
There is not enough disk space. Increase the amount of free space on the disk by deleting redundant files and so on. The amount of disk space can then be reduced by taking the following measures: • Setting a shorter storage period for performance data • Increasing the data collecting interval • Reducing the number of Performance Indicator items that are collected • Reducing the number of machines that are monitored
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
13 Failed to update the database. Details: Could not allocate new page for database 'RM_PerformanceDataBase2'. There are no more pages available in filegroup PRIMARY. Space can be
181
created by dropping objects, adding additional files, or allowing file growth.
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
29 Stopped Performance Monitoring Service because an irrecoverable error was detected. Details: Could not allocate new page for database 'RM_PerformanceDataBase2'. There are no more pages available in filegroup PRIMARY. Space can be created by dropping objects, adding additional files, or allowing file growth.
Action
The database size has reached 10 GB. System Monitor - Performance Monitoring Services uses SQL Server
2012 Express as its database engine. The maximum size supported by SQL Server 2012 Express is 10 GB.
Upgrade SQL Server Express to the SQL Server 2012 version, or delete data from the database by reducing the
storage period for the performance data with the Data Management Tool. The procedure is shown below.
1. Close the management console.
2. Start the System Monitor Data Management Tool. 3. Reduce the storage periods for the collected and summary data and then perform data summarization. After the data has been deleted, restore the performance data collecting interval to the original value. The amount of disk space can then be reduced by taking the following measures: • Setting a shorter storage period for performance data • Increasing the data collecting interval • Reducing the number of Performance Indicator items that are collected • Reducing the number of machines that are monitored
Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
67
The Database size has grow into CurrentSizeGB,larger than the warning size WarningSizeGB
Action
The database size has exceeded the specified warning value. Either change the warning value or delete data
from the database by reducing the storage period for the performance data with the Data Management Tool. The
procedure is shown below.
1. Close the management console.
2. Start the System Monitor Data Management Tool. 3. Reduce the storage periods for the collected and summary data and then perform data summarization. After the data has been deleted, restore the performance data collecting interval to the original value. The amount of disk space can then be reduced by taking the following measures: • Setting a shorter storage period for performance data • Increasing the data collecting interval • Reducing the number of Performance Indicator items that are collected • Reducing the number of machines that are monitored
182
9.9.3 Errors and Actions for Database Faults
The event logs relating to database references and updates, and the actions to take are listed below. Log Type Source ID Description
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
13 Failed to update the database. Details: General network error. Check your network documentation.
or
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
15 Failed to reference the database. Details: General network error. Check your network documentation.
System Monitor - Performance Monitoring Services
Error SystemMonitorPerformanceService
29 Stopped Performance Monitoring Service because an irrecoverable error was detected. Details: General network error. Check your network documentation.
Action
• Start the database service "MSSQL$SSCCMDB" if it has not started already. The database service starts automatically when the Performance Monitoring Service starts. The procedure for starting the service manually is as follows. When the instance name is changed from the default value, select “MSSQL$InstanceName” service name.
1. Start Services from the Administrative Tools control panel. 2. Select the service with the name "MSSQL$SSCCMDB" from the list of services that is displayed, and
double-click to open it. 3. Click Start in the General tab. 4. The database service will then start.
• Check whether there is a problem with the network connected to the management server. Check these points and resolve any problems, and then restart the Performance Monitoring Service. An error message may be output and the management console may terminate when the management console is started. If so, start the management console again.
9.9.4 Procedure for Backing up and Restoring the Database
Backing up and restoring database is executed using sqlcmd command offered in SQL Server.
<To back up>
(1) Close the management console by selecting Exit from the File menu of the management console main
window.
(2) Stop the Performance Monitoring Service.
- Start Services from the Administrative Tools control panel.
- From the list of services that is displayed, double-click the "System Monitor Performance Monitoring Service".
- Stop the Performance Monitor Service by clicking Stop in the General tab.
183
(3) Copy the following files to the backup destination:
- In the installation directory for System Monitor - Performance Monitoring Services
(Default value:%ProgramFiles%\NEC\SystemMonitorPerformance)
>> bin\rm_client.xml
>> bin\rm_service_init.xml
>> bin\rm_database.xml
(4) Execute the database backup command.
- Select All Programs from the Start menu, and then select Accessories and then Command Prompt.
- Specify the following command at the command line. In the example shown here, a backup file called
"sysmonbk.dat" is created in the \temp directory on the C drive.
> sqlcmd -E -S (local)\SSCCMDB -Q "backup database RM_PerformanceDatabase2 to disk =
'c:\temp\sysmonbk.dat' with init"
When the instance name is changed from the default value, specify “(local)\InstanceName”
<To restore>
(1) Close the management console by selecting Exit from the File menu of the management console main
window.
(2) Stop the Performance Monitoring Service.
- Start Services from the Administrative Tools control panel.
- From the list of services that is displayed, double-click the "System Monitor Performance Monitoring Service"
service.
- Stop the Performance Monitor Service by clicking Stop in the General tab.
(3) Restore the following files from the backup directory:
- In the installation directory for System Monitor - Performance Monitoring Services
(Default value:%ProgramFiles%\NEC\SystemMonitorPerformance)
>> bin\rm_client.xml
>> bin\rm_service_init.xml
>> bin\rm_database.xml
(4) Execute the database restore command.
- Select All Programs from the Start menu, and then select Accessories and then Command Prompt.
- Specify the following command at the command line. In the example shown here, a backup file called
"sysmonbk.dat" in the \temp directory on the C drive is restored.
> sqlcmd -E -S (local)\SSCCMDB -Q "restore database RM_PerformanceDatabase2 from disk =
'c:\temp\sysmonbk.dat' with replace"
When the instance name is changed from the default value, specify “(local)\InstanceName”.
184
9.10 Other
1. Perflib-related events recorded in the application log
Resolve the problem with the performance counter by contacting the developer of the application that owns
the performance counter (recorded as "***.dll" in the log) for which the problem occurred.
See Microsoft support technical information for details on performance library errors.
2. Warning errors for MSSQL$SSCCMDB:ID19011 recorded in the application log
The following event may be displayed in the event viewer when the management server starts:
• Type: Warning
• Source: MSSQL$SSCCMDB
• Event ID: 19011
This warning message results from the specification for SQL Server, and does not indicate a problem with
the execution of System Monitor - Performance Monitoring Services. See Microsoft support technical
information (ID: 303411) for details.
When the instance name is changed from the default value (SSCCMDB), “MSSQL$InstanceName” is
displayed.
3. Situations where logs are not recorded in the event log
In default setting for System Monitor - Performance Monitoring Services, previous logs are overwritten
when maximum log size (16MB) is reached. If the settings for the event viewer is changed to system default,
new logs may not be recorded when maximum log size is reached.
In this case, change the settings for the event viewer as follows:
(1) Start the Event Viewer from the Administrative Tools control panel.
(2) Right-click System Monitor - Performance Monitoring Services and select Properties. The System
Monitor - Performance Monitoring Services Properties window appears.
(3) In the General tab, change the Log size items.
4. Time difference between the management server and the console machine
When the management server is different from the management console machine, a time difference may
arise between the machines. System Monitor - Performance Monitoring Services adjusts the time
difference between the machines for graph display. If the time difference is greater than one hour, a
warning message is displayed.
185
Chapter 10 Notes
10.1 The Number of Monitored Machines
The maximum number of physical machines that can be specified as monitored targets with the single System
Monitor - Performance Monitoring Services is 300 units. The actual maximum number of units can be
controlled by the quantity of collected performance data and collection intervals. When large-scale system
configurations are handled with SystemProvisioning and the quantity of collected performance data is large
and the collection intervals are short, it is recommended that the configuration is divided by group units and
monitored using more than one System Monitor - Performance Monitoring Services or that the collection
interval is extended.
When the collection interval of performance data is 5 minutes, the maximum number of virtual machines that
can be specified as monitored targets with the single System Monitor - Performance Monitoring Services is 500
units.
186
10.2 Notes on Database
(1) Size of database
When 1 machine are to be monitored, there are 1 performance indicator, the collection interval is 1
minute, and the default storage period is to be used, then the data size can be estimated as follows.
Collected data: 3 (days) * 60 KB = 180 KB
5-minute summary data: 7 (days) * 30 KB = 210 KB
15-minute summary data: 30 (days) * 7 KB = 210 KB
1-hour summary data: 3 * 30 (days) * 2 KB = 180 KB
1-day summary data: 5 * 365 (days) * 0.1KB = 182.5 KB
Total size: (180 +210 + 210 + 180 + 182.5) (KB) * 1 (machines)
* 1 (performance indicator)
= 962.5 (KB) = approximately 1 (MB)
In SQL Server 2012 Express, used by System Monitor - Performance Monitoring Services as the database
engine, the maximum database size is 10 GB. Although a database file of 10 GB or more can be handled if
SQL Server 2012 Express is upgraded to SQL Server 2012, the system must be designed so that the size of
the performance data is less than or equal to 10 GB whenever SQL Server Express is being used.
Determine the number of computers to be monitored by a performance monitoring service based on the
number of performance data items to be collected, the collection interval, and the data storage period.
(2) Configuration of database
System Monitor Performance Monitoring Service cannot utilize the database installed on a remote server.
It is necessary to install the database on the System Monitor - Performance Monitoring Services
management server.
187
10.3 Notes on Graph Displays
(1) The period for historical displays
If the period for historical displays is too long for the plot interval, it may take some time for the graph to
be displayed in the management console. Reduce the display period, or increase the plot interval.
(2) Discontinued graph line
System Monitor - Performance Monitoring Services Management Console displays a graph as a "line" when
there is a plot data in the previous or next plot interval. When there is no plot data in both the previous and
next plot intervals, this plot data is displayed as only a "dot". The line of a displayed graph may be
discontinued when the data required to draw the graph is not partly present because the service is
restarted or a temporary access error occurs in a machine being monitored, or because the data collection
interval exceeds the specified plot interval.
(3) Graph display accuracy
If graph lines are made finer, the graph drawing precision may drop, and lines may become broken. If this
happens, make the lines thicker by modifying width in the Color and line tab of the Graph display setting
window.
(4) Data for graph display
System Monitor - Performance Monitoring Services determines the data used for graph display according
to the plot interval. If there is no data corresponding to the specified plot interval, the graph line for the
period containing no data is not displayed. See "1.4.2 Usage of Summary Data" for more details.
188
10.4 Notes on Linking to SystemProvisioning
The following three sections provide notes on using machines managed by SystemProvisioning as monitored
machines for System Monitor - Performance Monitoring Services.
(1) Applying changes in the system configuration information of SystemProvisioning to System Monitor -
Performance Monitoring Services
The SystemProvisioning configuration information can be manually or automatically applied to System
Monitor - Performance Monitoring Services. When this is automatic, System Monitor - Performance
Monitoring Services is transmitted to SystemProvisioning at fixed intervals and the changed configuration
information is automatically applied. Verify the following when enabling the application.
Verify the following if the application of system configuration information is failed.
• If the connection to SystemProvisioning is correct and communication fails between System Monitor -
Performance Monitoring Services and SystemProvisioning, confirm that the name of the
SystemProvisioning management server is correctly configured in the SystemProvisioning tab of the
Configuration window from the management console of System Monitor - Performance Monitoring
Services. When SystemProvisioning is clustered with EXPRESSCLUSTER, the virtual computer name
cannot be used as SystemProvisioning Management Server name. Using the virtual computer name can
cause a failure to connect SystemProvisioning.
In addition, if the Windows firewall function of the SystemProvisioning management server is valid,
configure the Windows firewall to accept exceptions by referring to Section, "2.6 SystemProvisioning
Connection Settings". When the automatic applying function is configured, consecutive errors generated
after the second time are not recorded to the event log or displayed as a message on the GUI.
• If the PVM Service has started. Start the PVM Service on the SystemProvisioning management server. If
the PVM service has been restarted during operations, restart the service of System Monitor -
Performance Monitoring Services (System Monitor Performance Monitoring Service) as well.
• Whether the service execution account is proper.
If the Performance Monitoring Service is not running on the SystemProvisioning management server, the
execution account of Performance Monitoring Service must have OS administrator authority for the
SystemProvisioning management server. If a proper account has not been set, change the service
execution account. See "2.5.2 How to Change the Performance Monitoring Service Execution Account" for
information on how to change the service execution account.
Apply the configuration information when the machine status returns to normal.
189
Confirm the following point if the performance collecting status of a target machine is changed to a paused
status from a normal status automatically.
• Whether the target machine status in SystemProvisioning is "Off" or "Unknown" in the power or OS
status, or "In-process" in the executing status.
The automatic configuration applying function checks the configuration status at regular intervals, so
SystemProvisioning configuration changes may not be applied immediately. Apply the configuration
manually or adjust the polling interval for automatic configuration applying.
If a monitored machine is running Linux, VMware ESX/ESXi, Citrix XenServer or KVM, System Monitor -
Performance Monitoring Services uses an IP address to connect to the monitored machine. If the IP address
of a monitored machine is not specified using the System Monitor - Performance Monitoring Services
management console then the machine name must be registered with either DNS or the “hosts” file on the
System Monitor - Performance Monitoring Services management server. The IP address settings for System
Monitor - Performance Monitoring Services monitored machines can be acquired when SystemProvisioning
configuration information is applied. If a machine is added using SystemProvisioning, IP address
information can also be acquired as additional machine information at the same time, which means that
operations can be automated. However, for some system configurations, there are cases where it is not
possible to connect to monitored machines using the IP addresses acquired, and so if a monitored machine
produces an error, either set the IP address using the Machine Settings window of the System Monitor -
Performance Monitoring Services management console or register the IP address with DNS or the “hosts”
file. If an IP address has been set up by using the Machine Settings window or as a result of applying
SystemProvisioning configuration information, the information registered in DNS or the “hosts” file will not
be looked up.
Then IP address has to be resolved as an IPv4 address.
If Change Group operation for the monitored machine is performed on SystemProvisioning, the historical
performance data is not inherited on System Monitor - Performance Monitoring Service because the
monitored machine is managed as another machine.
When the performance data collecting is enabled on the Performance Monitor settings for the group/model
on SystemProvisioning and then the application of SystemProvisioning configuration information is
performed on System Monitor - Performance Monitoring Service, the corresponding groups and machines
are register automatically on System Monitor - Performance Monitoring Service. When the performance
data collecting is disabled or "Do not Monitor" is selected as Profile Name on the Performance Monitor
settings for the group/model on SystemProvisioning and then the application of SystemProvisioning
configuration information is performed on System Monitor - Performance Monitoring Service, collecting
performance data for the corresponding groups and machines is paused on System Monitor - Performance
190
Monitoring Service. These group settings and machine settings are not deleted automatically from System
Monitor - Performance Monitoring Service. If necessary, delete manually via the System Monitor
Management Console.
(2) Configuration change of SystemProvisioning triggered by threshold monitoring
Load status abnormalities with monitored machines that are detected by threshold monitoring of System
Monitor - Performance Monitoring Services can be notified to SystemProvisioning. SystemProvisioning
receives these notifications and executes recovery process (such as adding machines) according to its
policies. Note the following points when configuring the notifications.
Conflicts with other products make notifications to SystemProvisioning.
Notifications to SystemProvisioning are made from multiple monitoring products. Care must be taken so
that these notifications do not duplicate System Monitor - Performance Monitoring Services notifications.
• Standard notifications (notifications from Server Management Manager)
Of all the events that can be used as SystemProvisioning standard notification events, do not set up
policies for events relating to performance status (such as “CPU load failure”) and policies for System
Monitor - Performance Monitoring Services's threshold monitoring notification for the same machine or
group. If both types of policies are set up, the recovery process specified in each policy may be executed for
the same machine at the same time.
If actions such as “Reboot machine” are set up for System Monitor - Performance Monitoring Services's
performance load abnormalities using SystemProvisioning policies, Server Management Manager may
raise a “Machine down” event while the machine is restarting. If a SystemProvisioning policy has been
set up for the “Machine down” event, a recovery process may be executed depending on the timing. Check
the SystemProvisioning log and restore the machine to an appropriate state.
• Custom notifications (notifications from optional products)
Monitoring products other than Server Management Manager make notifications to SystemProvisioning
using notification categories. Users can set up recovery processes for notification categories freely. For
these monitoring products, set up notification categories for the events that will trigger these recovery
processes. The same notification category can be assigned to more than one monitoring product, but the
roles for the notification categories should first be defined and then set up so that they do not conflict
when they are used.
As with the standard notifications setting, be careful when setting up policies so that recovery processes
actions are not duplicated.
Execution and timing of the configuration change action sequence
System Monitor - Performance Monitoring Services can be configured to detect abnormalities using the
frequency of threshold overruns with multiple checkpoints, to ensure that temporary load status is not
191
notified when notifying abnormalities of performance loads.
Similarly, if over-threshold status continues for an extended period of time, the timing for repeat
notifications can also be set up. Thresholds can be set in terms of upper-limits and lower-limits, for
“warning” or “critical”. Use these settings functions to design policies so that conflicts do not occur when
configuration changes are made. For example, appropriate settings are needed in the following kind of
situation.
• Settings were made for a group to add a machine when the threshold exceeds an upper-limit and to delete
a machine when the threshold exceeds a lower-limit, but machines were added and deleted repeatedly.
Review the settings for the upper- and lower-limit threshold values.
• Settings were made for a group to add a machine when the threshold exceeds an upper-limit, but
machines were added twice.
The performance load status had not recovered or high load may have been detected again before the first
machine addition completed. For the second case, review the timing for repeat notifications.
• If a certain group has been configured to move the VM when the threshold exceeds an upper-limit, the
VM was moved repeatedly.
Check the status of the destination VM server. Review the settings for the upper-limit threshold value
and, if necessary, review the operating method (consider moving the VM manually, for example).
SystemProvisioning discards notifications about machines for which actions are being executed. Logs are
written for both discarded events and received events, so check the logs.
With SystemProvisioning, different actions for notifications can be set up for each group/model. As a result,
different actions for the same customized notification category can be set up for different groups/models.
Take care to avoid conflicts when setting up actions for notifications.
The policy specified in the model settings will be given priority when SystemProvisioning received a
notification about the group which is set the full path to the model as a SystemProvisioning path.
The policy specified in the model settings will be given priority when SystemProvisioning received a
notification about the machine. The policy specified in the group settings will be enabled when no policy is
set in model settings. No recovery action will be executed when no policy is set in both group and model.
The SystemProvisioning path to be set in the group of System Monitor - Performance Monitoring Services
must be a full path to the model to enable the VM Optimization Function.
192
(3) Possible impacts made to performance status when SystemProvisioning configurations are changed
The performance status of the machines and groups being monitored by System Monitor - Performance
Monitoring Services may be affected when any changes to the system configuration are made, including
automatic changes executed by SystemProvisioning. The following situations may arise.
• When a machine fault occurs, performance data for the machine cannot be acquired. The performance
status of the group of the failed machine changes. And unexpected load may appear on that group.
• When the machine is in maintenance, performance data for the machine cannot be acquired correctly,
and the performance data for the group that includes the machine will not be valid.
With SystemProvisioning, configurations can change automatically, so if an unexpected load appears in the
performance status, check the log information for SystemProvisioning.
Also, if threshold monitoring notification has been set up for the group that includes the machine, a
recovery process (such as adding a machine to the group) may be executed by SystemProvisioning as a
result of a notification for the abnormal load status that occurred temporarily. In order to prevent
unwanted recovery processes from being executed, cancel settings for actions triggered by threshold
monitoring notification in situations where load can be anticipated in advance, such as when maintenance
is performed. In the case of an unexpected situation, such as faults, configure the threshold with an
allowance to ensure that an action is not generated at the same time, and specify a single action for the
same machine or group. If an unnecessary recovery process has been generated, confirm the log
information and return to the appropriate state.
193
10.5 Release of Resources
System Monitor - Performance Monitoring Services releases resources such as the libraries used to obtain
performance data, at AM 3:00 every day. The release of such resources is normally completed after several
minutes. Note that the collection of performance data and the monitoring of the threshold are stopped
during the release.
194
10.6 Notes on Upgrading/Updating
10.6.1 Changing the Setting of SystemProvisioning Configuration Path
The path of SystemProvisioning configuration information must be a full path to the group/model. If a path to
a SystemProvisioning category, which is inherited from an earlier version, is specified in the group settings of
System Monitor - Performance Monitoring Services, change the setting to indicate the path to the
group/model.
10.6.2 Additional Setting for Monitoring Linux Machine with Multipath Configuration
An additional setting is necessary if a monitored Linux machines has a multipath configuration using NEC
iStorage StoragePathSavior. Change the setting using the following procedure if the update module for
SigmaSystemCenter 2.1 (SSC0201-0006-update2 or SSC0201-0015-update3) was applied to your system in
past.
1. Stop the Management Console.
2. Stop the System Monitor Performance Monitoring Service.
3. Open the configuration file (\bin\rm_service_init.xml) on the installation directory with text editor after
backing up it. (Default installation directory: %ProgramFiles%\NEC\SystemMonitorPerformance)
4. Change the setting of the LinuxDiskDeviceNameRegex tag as shown below:
5. Over write the configuration file and start the System Monitor Performance Monitoring Service.
10.6.3 Changing the Performance Indicator Title
If a title of the built-in performance indicator had been modified from the default title in SigmaSystemCenter
2.1 or earlier, the title will be changed automatically to the default title during upgrading procedure.
Reconfigure the graph settings of the management console with a default performance indicator title.
10.6.4 Changing the Error Code Format for User Scripts
The format "@errorcode number," which is used in SigmaSystemCenter 3.2 or earlier, is deprecated in
SigmaSystemCenter 3.3. SigmaSystemCenter 3.3 allows for you to use only the format "errorcode=number "
as an error code format. If the format "@errorcode number" is used in a script to return an error code, change
the format to "errorcode=number."
<LinuxDiskDeviceNameRegex>(?:hd|sd|xvd|dd)[a-z]+</LinuxDiskDeviceNameRegex>
195
10.7 Issues When Other Products to Monitor the Windows Monitored Machine Coexist with System Monitor - Performance Monitoring Services
Adding to System Monitor - Performance Monitoring Services, if other performance monitoring products
monitor Windows monitored machines, the "Category does not exist" error might occur while collecting the
performance data.
System Monitor - Performance Monitoring Services uses the Windows performance monitor feature to obtain
the performance data from the Windows monitored machine. This issue occurs other monitoring products also
use the Windows performance monitor feature as well as System Monitor - Performance Monitoring Services.
Use either System Monitor - Performance Monitoring Services or the other product to monitor the Windows
monitored machine.
196
Appendix
A. How performance values for groups are calculated
Usage rates (group performance) for a group made up of multiple machines can be calculated as either the
absolute performance values of each machine or as the weighted average of the usage rate for each machine,
using performance ratios as coefficients. The following model formula is generally suitable for calculating group
usage rates.
For CPU usage rates, the group usage rate can generally be calculated by using the published benchmark value
for αi. Graphs that show the performance status of a group while taking into account the ability of the machines
that make up the group can be displayed by setting this αi value in the Weight field in the Machine Settings
window for the System Monitor - Performance Monitoring Services management console and selecting
Weighted average value as the statistic in the Graph display object settings window. However, care is required
if SMP or Hyper-Threading and so forth is being applied, as the trends are different to the benchmark values.
(Performance improves by 50 to 100% for SMP and by 0 to 60% for Hyper-Threading (30% on average).) Further
errors arise depending on the applications executing on the monitored machines. Pre-evaluations are needed in
order to measure usage rates accurately.
For disk usage rates, the trends are different to the benchmark values, and so αi must be derived accurately
using pre-evaluations. However, if disks are generally under medium to heavy load, the performance of each
disk can be regarded as almost the same (αi is constant), and the measurement error is about ±15% even the
simple average of the usage rate for each machine is taken as the usage rate of the group. Disk usage rates can
be specified using the Windows custom definitions of the Performance Indicator settings.
∑∑=
i i
ii i α
αρρ
The views expressed here regarding the model formula and CPU and disk usage rates have been derived
from test results using IIS/ASP. For actual operations, it is recommended that you set values based on
pre-evaluations taking into account such factors as the characteristics of the applications being used.
ρ: Group usage rate
ρi: Machine usage rate
αi: Performance value relative to other machines (performance ratio)