NEC Express5800/320Ma User manual

NEC Solutions (America), Inc.
NR007W
Express5800/320Ma:
Software Availability Manager
User’s Guide

Manual Name: Express5800/320Ma: Software Availability Manager
Part Number: NR007W
Express5800/320Ma Software Release Number: 4.1.0
Publication Date: January 2006
NEC Solutions (America), Inc.
10850 Gold Center Drive, Suite 200
Rancho Cordova, CA 95670
© 2006 NEC Solutions (America), Inc. All rights reserved.
Notice
The information contained in this document is subject to change without notice.
UNLESS EXPRESSLY SET FORTH IN A WRITTEN AGREEMENT SIGNED BY AN AUTHORIZED REPRESENTATIVE
OF NEC, NEC MAKES NO WARRANTY OR REPRESENTATION OF ANY KIND WITH RESPECT TO THE
INFORMATION CONTAINED HEREIN, INCLUDING WARRANTY OF MERCHANTABILITY AND FITNESS FOR A
PURPOSE. NEC assumes no responsibility or obligation of any kind for any errors contained herein or in connection with
the furnishing, performance, or use of this document.
Software described in NEC (a) is the property of NEC and/or its licensees, (b) is furnished only under license, and (c) may
be copied or used only as expressly permitted under the terms of the license.
NEC documentation describes all supported features of the user interfaces and the application programming interfaces
(API) developed by NEC and/or its licensees. Any undocumented features of these interfaces are intended solely for use
by NEC personnel and are subject to change without warning.
This document is protected by copyright. All rights are reserved. No part of this document may be copied, reproduced, or
translated, either mechanically or electronically, without the prior written consent of NEC Solutions (America), Inc.
The NEC Solutions (America), Inc. logo, Express5800/320Ma, and the Express5800/320Ma logo, are trademarks of NEC
Solutions (America), Inc. Software Availability Manager and ActiveService Network are trademarks of Stratus
Technologies Bermuda, Ltd. All other trademarks and trade names are the property of their respective owners.

Contents iii
Contents
Preface vii
1. Overview of the Software Availability Manager 1-1
Software Availability Manager 1-1
Software Availability Manager Functions 1-2
Monitoring Instances and Events 1-2
SAM Instances 1-3
Using Email and Executing Programs 1-4
Event Log Messages from SAM 1-4
2. SAM Settings Reference 2-1
Express5800/320Ma Monitored Instances Tab 2-1
Instance Parameters 2-2
Alert Thresholds 2-5
Warnings and Alarms 2-5
Threshold Units 2-6
How SAM Scans an Instance 2-7
3. Configuring the Software Availability Manager 3-1
Configuring Email 3-1
Email User Setup 3-2
Email Alert Setup 3-3
Alert Email Notification 3-3
Default Email Alert Notification 3-3
Nondefault Alert Notification 3-4
Associating a Nondefault Alert ID with an Instance 3-5
Deleting an Alert ID 3-6
Sending No Notification 3-7
Testing Email 3-7
Setting Up SAM to Run a Program 3-8
Changing Default Values of the Instance Parameters 3-8

Contents
iv Express5800/320Ma: Software Availability Manager
SAM Instances 3-9
Context Switches (Per Second) 3-9
CPU Usage 3-10
Disk Free Space 3-10
Event Log Messages 3-11
Memory Available 3-11
Page File Usage 3-11
4. Sample Configuration 4-1
Sample Configuration 4-1
Displaying the SAM Interface 4-2
Configuring Context Switches (Per Second) 4-2
Configuring CPU Usage 4-3
Configuring Disk Free Space 4-4
Configuring Event Log Messages 4-5
Configuring Memory Available 4-6
Configure Page File Usage 4-7
Making Changes Effective 4-8
Glossary Glossary-1
Index Index-1

Figures v
Figures
Figure 2-1. Enabling Alerts 2-2
Figure 2-2. Scan Poll Rate, Email Alert ID, and Program to Launch 2-2
Figure 2-3. Scan-Related Parameters 2-3
Figure 2-4. Threshold Setup 2-6
Figure 2-5. Scanning Diagram, Part 1 2-8
Figure 2-6. Scanning Diagram, Part 2 2-9
Figure 3-1. Sample Default Email Alert 3-4
Figure 3-2. Accepting Changes to the Configuration 3-6
Figure 3-3. Email Configuration Error 3-7
Figure 4-1. Changing Units of Measurement 4-5
Figure 4-2. Accepting Changes to the Configuration 4-8

Tables vi
Tables
Table 4-1. SAM Parameters 4-1

Preface vii
Preface
Purpose of This Manual
TheExpress5800/320Ma:SoftwareAvailability Managerexplainshowto configureand
use ftServer Software Availability Manager, an application that monitors the
performanceofcritical Express5800/320Masystemcomponents andthestatusofuser
applications.
Audience
This manual is intended for personnel who administer or monitor Express5800/320Ma
systems.
Notation Conventions
This document uses the notation conventions described in this section.
Warnings, Cautions, and Notes
Warnings, cautions, and notes provide special information and have the following
meanings:
WARNING
!A warning indicates a situation where failure to take
oravoid aspecifiedactioncouldcausebodilyharmor
loss of life.
CAUTION
!A caution indicates a situation where failure to take or
avoidaspecifiedactioncould damagea hardware device,
program, system, or data.
NOTE
Anote providesimportantinformation abouttheoperation
of an Express5800/320Ma system.

Preface
viii Express5800/320Ma: Software Availability Manager
Typographical Conventions
The following typographical conventions are used in Express5800/320Ma documents:
•The bold font emphasizes words in text or indicates text that you type, the name of
a screen object, or the name of a programming element. For example:
Before handling or replacing the clock card, make sure that you are properly
grounded by using a grounded wrist strap.
In the System Properties dialog box, click the Hardware tab.
Call the RegisterDeviceNotification function.
•Theitalic fontintroducesnew terms andindicatesprogramming andcommand-line
arguments that the user defines. For example:
Many hardware components are customer-replaceable units (CRUs), which
can be replaced on-site by system administrators with minimal training or tools.
copy filename1 filename2
Pass a pointer for the NotificationFilter parameter
•The monospace font indicates sample program code and output, including
message text. For example:
#include <iostream.h>
The operation completed successfully.
Getting Help
If you have a technical question about Express5800/320Ma hardware or software, try
these online resources first:
•Online support from NEC Technical Support. You can find the latest technical
information about an Express5800/320Ma through online product support at the
NEC Technical Support Web site:
http://support.necsam.com/servers/
•Online product support for Microsoft®products. Your primary source for
support is the computer manufacturer who provided your software, or an
authorized Microsoft Support Provider. You can also find the latest technical
informationaboutMicrosoftWindows®andotherMicrosoftproductsthroughonline
product support at the Microsoft Help and Support Web site:
http://support.microsoft.com/

Preface
Preface ix
If you areunable to resolve your questions with the help available atthese online sites,
and the Express5800/320Ma system is covered by a service agreement, please
contact NEC Technical Support (866-269-1239).
Notices
•All regulatory notices are provided in the site planning guide for your system.
•Although this guide documents modem functionality, modems are not available for
all systems. Ask your sales representative about modem availability.
•ActiveService Network (ASN) is not currently available but may be ordered in the
future.

Preface
xExpress5800/320Ma: Software Availability Manager

Overview of the Software Availability Manager 1-1
Chapter 1
Overview of the Software Availability
Manager1-
The ftServer Software Availability Manager (SAM) Service monitors system
performanceand user-applicationavailability.This chapterpresents an overviewof the
SAM service. It includes the following topics:
•“Software Availability Manager”
•“Event Log Messages from SAM”
•“Event Log Messages from SAM”
Software Availability Manager
The Software Availability Manager consists of a Windows service that runs at system
startup and a snap-in that runs within a Microsoft Management Console (MMC). The
SAM Service monitors the performance of critical Express5800/320Ma system
components and the status of user applications.
You can configure the SAM snap-in to send email or execute a program based on the
status of applications, configured performance thresholds, or messages sent to the
Event Log.
SAMis oneoftwoNEC-suppliedMMCsnap-ins.Theotheris theftServerManagement
Console (ftSMC) snap-in, which is documented in the system administrator’s guide.
SAM and ftSMC both monitor system functionality. Most of the elements (called
instances) that SAM monitors relate to system performance, though it also detects
user-application failures and monitors Event Log entries for warning and error
messages. On the other hand, ftSMC monitorssystem hardware and the device drivers
for the ftServer system.
SAM and ftSMC both report system information to system administrators.

Software Availability Manager
1-2 Express5800/320Ma: Software Availability Manager
Software Availability Manager Functions
SAM has the following functions:
•Monitor user applications to detect when they fail.
•Monitor the performance of critical system components—such as disk, CPU, and
memory—to detect performance problems before they cause a crash or data loss.
•Send messages to the Windows Event Log and notify users, by email, of certain
system-performance conditions.
•Execute user-specified programs in response to configured alert conditions.
Monitoring Instances and Events
SAM monitors instances and events.
Instances are the system performance elements that you specify to be monitored; for
instance, CPU usage, disk free space, and so on. The list of instances you can monitor
appearsontheExpress5800/320MaMonitoredInstancestaboftheSAMsnap-in.(See
“SAMInstances” onpage 1-3foracompletelistofinstancesthatyoucanmonitor).You
specify which instances to monitor by configuring SAM in the MMC console.
An event, in the context of SAM, is a state change in a monitored instance. For
example, available disk space falling below specified limits is an event. You can
configure the instances to generate a warning or alarm and trigger the execution of a
program when the instance reaches a certain value.
For example, you can configure SAM to:
•Send an email warning (less severe) or alarm (more severe) to members of a
defined email list when available disk space on the system falls to a specified
percentage.
•Send email notification to members of a defined email list when the available
system memory falls below a specified amount.
•Execute a program when some specified threshold value is reached.

Software Availability Manager
Overview of the Software Availability Manager 1-3
SAM Instances
You can monitor and configure the following instances:
•Context Switches (Per Second). The number of switches per second from one
thread to another. A context switch occurs when the processor stops processing
one thread and starts processing another. Context switches per second is a
measure of how the microkernel is distributing the CPU processing of threads and
processes on the system.
•CPUUsage.Thepercentageoftimethattheprocessorisbusyexecutinganon-idle
thread. CPU usage measures how much work the CPU is doing and which
processes are using the CPU’s time. Sustained high CPU usage can degrade
system performance.
•Disk Free Space. The ratio of the free disk space to total usable space, per each
logical disk on a system. The threshold values supplied for this instance apply to
each logical disk in the system. If any logical disk's free space falls below the
configured threshold, for instance, 5%, SAM would send email notification
(assuming you configured the instance to do so).
•Event Log:
– Application Error Messages
– Application Warning Messages
– System Error Messages
– System Warning Messages
The Event Log instances detect when the Event Log receives a System or
Application error or warning message.
•Memory Available. The size in megabytes (MB) of the virtual memory currently
available in the system. This quantity combines the size of RAM memory plus the
size of the system page file. Running out of virtual memory can cause system
instability or even application failures.
•Page File Usage. The amount of page file currently in use. This can help determine
whether the page file is too small for your system.

Event Log Messages from SAM
1-4 Express5800/320Ma: Software Availability Manager
Using Email and Executing Programs
To send email notification, or to have specified programs run automatically when
certain threshold values are reached, you must first configure the SAM email facility.
See “Configuring Email” on page 3-1 for instructions on how to configure the SAM
email facility.
Event Log Messages from SAM
Like any Windows service, SAM sends the following kinds of messages to the Event
Log:
•Informational messages, such as service started and service stopped.
•Warning messages. These are reports of the email alerts that SAM sent in
response to configured threshold specifications.
•Error messages. These are SAM-based errors, such as invalid configurations, or
email failures caused by invalid addresses, and so on.

SAM Settings Reference 2-1
Chapter 2
SAM Settings Reference2-
ThischaptersummarizesthesettingsoftheSoftwareAvailabilityManagerconfigurable
instances and explains the warning and alarm thresholds. It presents the following
topics:
•“Express5800/320Ma Monitored Instances Tab”
•“Instance Parameters”
•“Alert Thresholds”
•“How SAM Scans an Instance”
Express5800/320Ma Monitored Instances Tab
You view and configure the SAM instances and their threshold values on the
Express5800/320Ma Monitored Instances tab.
To display the Express5800/320Ma Monitored Instances tab
1. On the Express5800/320Ma desktop, click the ftServer version Management
Tools icon.
2. Right-click the ftServer Software Availability Manager icon, then click
Properties.
TherearetwononstandardbuttonsontheExpress5800/320Ma Monitored Instances
tab:
•Reset All restores all instances to their default settings.
•Use Default reselects the default email alert for the selected instance.

Instance Parameters
2-2 Express5800/320Ma: Software Availability Manager
Instance Parameters
Instances are the system performance elements that the Express5800/320Ma SAM
monitors; for example, CPU usage, free disk space, and so on. A description of the
instances and related settings follows.
The Enable alerts for selected instance setting enables warnings and alarms to be
generated when the monitored component highlighted in the Select Instance menu
exceeds its thresholds.
Figure 2-1 shows alerts enabled for the CPU Usage instance.
Figure 2-1. Enabling Alerts
NOTE
TheEnable alerts for selected instance check boxmust
be selected to enable alert notification for the selected
instance.
If you enable alerts for an instance, you must select at least one threshold value.
Figure 2-2 shows the Scan Poll Rate, Email Alert ID, and Program Launch fields.
Figure 2-2. Scan Poll Rate, Email Alert ID, and Program to Launch
Scan Poll Rate (in seconds). The interval, in seconds, at which SAM will scan (poll)
aninstance todetectapossible thresholdviolation.Ifthis rateis300,for example,SAM
scans the instance’s value once every 300 seconds, until detecting a threshold
violation.
However, whether a threshold violation actually occurs also depends on whether Scan
retry is enabled or disabled.

Instance Parameters
SAM Settings Reference 2-3
If an instance’s value exceeds its threshold:
•and Scan retry is disabled, a threshold violation has occurred. In this case, SAM
does whatever it has been configured to do in the event of a threshold violation. For
instance, it might send email notification or run a program.
•and Scan retry is enabled, there is not yet a threshold violation, and SAM
continues to scan at the Scan retry poll rate (in seconds).
NOTE
See “How SAM Scans an Instance” on page 2-7 for more
information about the scan-related parameters.
The maximum value for this parameter varies per instance.
Email Alert ID. A user-defined string that identifies the alert to be sent. This box must
contain either the value “Default,” if the associated instance will send the default alert,
or an identifier you created in the Email Alert Setup tab.
Creating a nondefault alert ID and associating it with an instance is described in
“Nondefault Alert Notification” on page 3-4.
The string may be of any length. “Default” is the default setting.
Program to launch (optional). The full path name of the program you want to launch
(execute) in the event of a threshold violation. The program that runs must not be
interactive, as no window is displayed. Also, the program must exit by itself. This box
is left blank by default.
Figure 2-3 shows some of the parameters related to scanning. See “How SAM Scans
an Instance” on page 2-7 for an explanation of how SAM uses the scanning
parameters.
Figure 2-3. Scan-Related Parameters

Instance Parameters
2-4 Express5800/320Ma: Software Availability Manager
Scan retry. A checkbox that enables you to use Number of scan retries and Scan
retry poll rate (in seconds), which work in tandem.
You must check Scan retry to be able to specify a value for Number of scan retries and
Scan retry poll rate.
Number of scan retries. The number of times to retry the scan at the Scan retry poll
rate interval.
When Number of scan retries is enabled, if an instance’s monitored value exceeds its
threshold, SAM continues to scan the instance at the Scan poll rate. If the Number of
scan retries value is reached and the instance’s value still exceeds its threshold, a
threshold violation occurs. SAM will then generate an alert.
If you disable Number of scan retries (by disabling Scan retry), the first detected
threshold violation will generate an alert.
The range of values for this parameter varies per instance.
Scan retry poll rate (in seconds). The interval at which SAM makes subsequent
scans after detecting a first threshold violation. The subsequent scans are made to
verify that the threshold violation is not temporary (a spike).
You must enable Scan retry in order to use Scan retry poll rate.
The maximum value for this parameter varies per instance.
Scan idle. A check box that enables you to use Time between alerts (in seconds).
You must enable Scan idle to be able to specify a value for Time between alerts.
Time between alerts (in seconds). The number of seconds during which an instance
is not scanned after an alert has occurred. The Time between alerts is an idle period
for scanning.
Enabling this parameter prevents multiple alerts from being issued during the specified
period. This minimizes the number of alerts triggered by an instance that remains
outside of the normal range.
You must enable Scan idle in order to use Time between alerts.
The maximum value for this parameter is 72 hours (259200 seconds). A more typical
value is 3600 (1 hour).

Alert Thresholds
SAM Settings Reference 2-5
Alert Thresholds
SAM monitors instances by scanning, or polling, the current values of each instance.
Each instance is configured with default alert-threshold values expressed in an
appropriate unit of measure. When an instance exceeds its threshold, it enters one of
two possible conditions: warning or alarm.
NOTE
You can change any of the defaults to values more
suitable to your environment. See “Changing Default
Values of the Instance Parameters” on page 3-8 for more
information.
Warnings and Alarms
Alarms are more urgent than warnings. The thresholds that trigger warnings and
alarms can be either high or low, depending on the instance being monitored. For
instance, you should configure both low disk space, on the one hand, and a too-high
rate of CPU usage, on the other, to generate alarms.
Figure 2-4 shows the threshold settings (as well as the Units parameters) on the
Express5800/320Ma Monitored Instances tab. The figure shows High Threshold
Warning and High Threshold Alarm as they might be set for the CPU Usage instance.
A CPU Usage level of 80% causes a High Threshold Warning; a 90% usage level
causes a High Threshold alarm.
The Disk Free Space instance uses low thresholds. You might have the system
generate a warning when it detects a Disk Free Space value of 20%, and an alarm
when it detects only 10%.
The threshold values are:
•Low Threshold Warning. The threshold is a low value, causing a warning alert
notification. The next lower threshold causes an alarm.
•High Threshold Warning. The threshold is a high value, causing a warning alert
notification.The next higher threshold causes an alarm.
•Low Threshold Alarm. The threshold is a low value, causing an alarm.
•High Threshold Alarm. The threshold is a high value, causing an alarm.
Except for the System Event Log, which has no thresholds, you configure instances
with either a low or high condition, as appropriate. Also, you have the option of defining
only one condition, for example, Low Threshold Alarm, without defining a Low
Threshold Warning.

Alert Thresholds
2-6 Express5800/320Ma: Software Availability Manager
(An instance can also theoretically be configured with both high and low thresholds,
each with warning and alarm conditions, totaling four settings, though there is currently
no instance that needs this.)
Figure 2-4. Threshold Setup
Threshold Units
Each instance includes a Units item (shown in Figure 2-4) among its parameters. The
system automatically sets this parameter.
The Units parameter determines the unit of measure being scanned and interpreted as
high or low. Unit values are:
•Percent, a value, from 1 to 100, of a component’s capacity.
•Megabytes, units of 1,048,576 bytes.
•Count, for discrete quantity, like the number of context switches per second.
•Message, for those instances whose only action is to receive messages, like
Application Fault Monitor.
For example, the CPU Usage instance’s Units parameter is defined as Percent.
Therefore, a High Threshold Alarm value of 90 means that SAM will produce an alarm
when the CPU is operating at 90% of capacity. On the other hand, the Memory
Available instance’s Units parameter is set to Megabytes. Therefore, a Low Threshold
Alarm value of 4 means that the system will produce a Low Threshold Alarm alert when
only 4 MB of memory are available.