HP OpenVMS Systems Documentation

HP OpenVMS Availability Manager User's Guide

This guide explains how to use HP Availability Manager software to detect and correct system availability problems.

Revision/Update Information: This guide supersedes the HP OpenVMS Availability Manager User's Guide, Version 3.1.

Operating System: Data Analyzer:
Windows 2000 SP 4 or higher;
Windows XP SP 2;
OpenVMS Alpha Versions 8.2 and 8.3;
OpenVMS I64 Versions 8.3 and 8.3-1H1

Data Collector:
OpenVMS VAX Version 6.2 and 7.3;
OpenVMS Alpha Versions 8.2 and 8.3;
OpenVMS I64 Versions 8.3 and 8.3-1H1

Software Version: HP Availability Manager Version 3.1

Hewlett-Packard Company Palo Alto, California

© Copyright 2010 Hewlett-Packard Development Company, L.P.

Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor's standard commercial license.

The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein.

Intel and Itanium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation.

Java is a US trademark of Sun Microsystems, Inc.

Printed in the US

ZK6552

Intended Audience

This guide is intended for system managers who install and use HP Availability Manager software. It is assumed that the system managers who use this product are familiar with Microsoft Windows terms and functions.

Document Structure

This guide contains the following chapters and appendixes:

• Chapter 1 provides an overview of Availability Manager software, including security features.
• Chapter 2 tells how to set up and configure the Data Analyzer and Data server, how to start the Data Server and Data Analyzer, use the main System Overview window, select a group of nodes and individual nodes, and use online help.
• Chapter 3 tells how to select nodes and display node data; it also explains what node data is.
• Chapter 4 tells how to display OpenVMS Cluster summary and detailed data; it also explains what cluster data is.
• Chapter 5 tells how to display and interpret events.
• Chapter 6 tells how to take a variety of corrective actions, called fixes, to improve system availability.
• Chapter 7 describes the tasks you can perform to filter, select, and customize the display of data and events.
• Appendix A contains a table of CPU process states that are referred to in Section 3.2.2.4 and in Section 3.3.1.
• Appendix B contains a table of OpenVMS and Windows events that can be displayed in the Event pane discussed in Chapter 5.
• Appendix C describes the events that can be signaled for each type of OpenVMS data that is collected.

Related Documents

The following manuals provide additional information:

• HP OpenVMS System Manager's Manual describes tasks for managing an OpenVMS system. It also describes installing a product with the POLYCENTER Software Installation Utility.
• HP OpenVMS System Management Utilities Reference Manual describes utilities you can use to manage an OpenVMS system.
• HP OpenVMS Programming Concepts Manual explains OpenVMS lock management concepts.

For additional information about HP OpenVMS products and services, see:

http://www.hp.com/go/openvms 

Reader's Comments

HP welcomes your comments on this manual. Please send your comments or suggestions to:

openvmsdoc@hp.com 

How to Order Additional Documentation

For information about how to order additional documentation, see:

http://www.hp.com/go/openvms/doc/order 

Conventions

The following conventions are used in this guide:

This chapter answers the following questions:

• What is the HP Availability Manager?
• How does the Availability Manager work?
• How does the Availability Manager maintain security?
• How does the Availability Manager identify possible performance problems?

The HP Availability Manager is a system management tool that allows you to monitor, from an OpenVMS or Windows node, one or more OpenVMS nodes on an extended local area network (LAN).

The Availability Manager helps system managers and analysts target a specific node or process for detailed analysis. This tool collects system and process data from multiple OpenVMS nodes simultaneously, analyzes the data, and displays the output using a graphical user interface (GUI).

Features and Benefits

The Availability Manager offers many features that can help system managers improve the availability, accessibility, and performance of OpenVMS nodes and clusters.

Figure 1-1 is an example of the initial System Overview window of the Availability Manager.

Figure 1-1 System Overview Window

The System Overview window is divided into the following sections:

• In the upper section of the display is a list of user-defined groups and a list of nodes in each group. You can compress the display to only the name of a group by clicking the handle preceding the group name. The summary group line remains, showing the collected information for all the nodes in the group.
  If a node name displays a red icon, you can hold the cursor over the icon, the node name, or the number in the Events column to display a tooltip explaining what the problem is; for example, for the node DBGAVC, the following message is displayed:

  HIHRDP, high hard page fault rate

  
  This section of the window is called the Group/Node pane.
• In the lower section of the window events are posted, alerting you to possible problems on your system. The items on the pane vary, depending on the severity of the problem: the most severe problems are displayed first. This section of the window is called the Event pane.

The Availability Manager consists of the following parts:

• Data Collector
  Runs on OpenVMS nodes and has two purposes:
  • accepts requests for data from a Data Analyzer
  • allows a Data Analyzer or Data Server to communicate with other Data Collectors
• Data Analyzer
  Runs on an OpenVMS or Windows node. It displays collected data in an easy-to-use graphic user interface (GUI).
• Data Server
  Runs on an OpenVMS or Windows node. It allows the Data Collector and Data Analyzer to communicate over a wide area network (WAN) using the Internet Protocol (IP) suite.

The next two sections describe how these parts work together on an extended LAN and on a WAN.

1.2.1 Data Analyzer and Data Collector on the Same Extended LAN

The Data Analyzer and Data Collector communicate over an extended LAN using an IEEE 802.3 Extended Packet format protocol. Once a connection between the Data Analyzer and the Data Collector is established, the Data Analyzer instructs the Data Collector to gather specific system and process data.

Although the Data Analyzer can be run on a member of a monitored cluster, it is typically run on a system that is not a member of a monitored cluster. Because of this, the Data Analyzer does not hang if the cluster hangs.

When the Data Analyzer and Data Collectors reside on the same extended LAN, they can communicate directly with each other. Restrictions on this direct communication setup are as follows:

• Only one Data Analyzer can run on a system at one time.
• Communication between the Data Analyzer and Data Collectors is not routable in an IP network.

        Protocol ID:        08-00-2B-80-48 
        Multicast Address:  09-00-2B-02-01-09 

Figure 1-2 shows a possible configuration of nodes running Data Analyzers and Data Collectors on an extended LAN.

Figure 1-2 Availability Manager Node Configuration for an Extended LAN

In Figure 1-2, the Data Analyzer can monitor nodes A, B, and C across the network. The password on node D does not match the password of the Data Analyzer; therefore, the Data Analyzer cannot monitor node D.

For information about password security, see Section 1.3.

Requesting and Receiving Information over an Extended LAN

After installing the Availability Manager software, you can begin to request information from Data Collectors on one or more nodes.

Requesting and receiving information requires the Availability Manager to perform several steps, which are shown in Figure 1-3 and explained in the text following the figure.

Figure 1-3 Requesting and Receiving Information over an Extended LAN

The following steps correspond to the numbers in Figure 1-3.

1. The Data Analyzer passes a user's request for data to the driver on the Data Analyzer node:
   • On Windows systems, the Windows driver is part of the Windows kit.
   • On OpenVMS systems, the OpenVMS driver is called the Data Collector driver and is included in the Data Collector kit. This is the same driver that is on the Data Collector node.
2. The driver on the Data Analyzer transmits the request across the network to the driver on the Data Collector node.
3. The driver on the Data Collector transmits the requested information as data over the network to the driver on the Data Analyzer node.
4. The driver on the Data Analyzer node passes the data to the Data Analyzer, which displays the data.

In step 4, the Data Analyzer also checks the data against various thresholds and conditions, and posts events if the thresholds are exceeded or the conditions met. The following section explains how data analysis and event detection work.

Data Collector Notes

There are some characteristics to note about the Data Collector drivers on OpenVMS and Windows.

• The Data Collector on a Data Collector node can collect data for more than one Data Analyzer node at the same time.
• The Data Collector driver on an OpenVMS Data Analyzer node can only support one Data Analyzer at a time.
• The Data Collector driver on a Windows Data Analyzer node can only support one Data Analyzer connection to a network adapter at a time.

The Data Analyzer can communicate only with Data Collectors that are on an extended LAN. (LANs are usually limited to a building or even just to a computer room.) However, you might need to run a Data Analyzer on a node that is not part of an extended LAN---for example, from home or at another site. To do this, you must add a Data Server node to your extended LAN.

The purpose of the Data Server node is to relay data between the Data Analyzer and Data Collectors. The Data Server formats data for transport to and from the Data Analyzer over a WAN.

Figure 1-4 shows an example of adding a Data Server and WAN connection to Figure 1-2.

Figure 1-4 Availability Manager Node Configuration for a WAN

In Figure 1-4, the Data Analyzer monitors Data Collector nodes by passing data through the Data Server. When you start the Data Analyzer, you direct it to connect to the Data Server over the WAN. Once the connection is established, the Data Analyzer can connect to Data Collectors through the Data Server and start collecting data.

Requesting and Receiving Information over a WAN

After installing the Availability Manager software, you can begin to request information from Data Collectors on one or more nodes.

Requesting and receiving information requires the Availability Manager to perform several steps, which are shown in Figure 1-5 and explained in the text following the figure.

Figure 1-5 Requesting and Receiving Information Over a WAN

The following steps correspond to the numbers in Figure 1-5.

1. The Data Analyzer passes a user's request for data to the IP socket connection on the Data Analyzer node.
2. Using a secure socket, the IP socket transmits the request to the IP socket connection on the Data Server node.
3. The IP socket on the Data Server node passes the request to the Data Server.
4. The Data Server passes the request to the IP socket on the Data Server node:
   • On Windows systems, the Windows driver is part of the Windows kit.
   • On OpenVMS systems, the OpenVMS driver is called the Data Collector driver and is included in the Data Collector kit. This is the same driver that is on the Data Collector node.
5. The driver on the Data Server transmits the request across the network to the driver on the Data Collector node.
6. The driver on the Data Collector transmits the requested information as data over the network to the driver on the Data Server node.
7. The driver on the Data Server node passes the data to the Data Server.
8. The Data Server passes the data to the IP socket connection.
9. The IP socket on the Data Server node transmits the data to the IP socket on the Data Analyzer node.
10. The IP socket on the Data Analyzer node passes the data to the Data Analyzer, which displays the data.

In step 10, the Data Analyzer also checks the data for any events that need to be posted. The following section explains how data analysis and event detection work.

The Availability Manager uses passwords to maintain security. Passwords are eight alphanumeric characters long. The Data Analyzer stores passwords in its customization file. On OpenVMS Data Collector nodes, passwords are part of a three-part security code called a security triplet.

The following sections explain these security methods further.

1.3.1 Data Analyzer Password Security

For monitoring to take place, the password on a Data Analyzer node must match the password section of the security triplet on each OpenVMS Data Collector node. OpenVMS Data Collectors also impose other security measures, which are explained in Section 1.3.2. This password match is used whether or not a Data Server is involved in the connection between the Data Analyzer and the Data Collector.

Figure 1-6 illustrates how you can use passwords to limit access to node information.

Figure 1-6 Availability Manager Password Matching

As shown in Figure 1-6, the Testing Department's Data Analyzer, whose password is HOMERUNS, can access only OpenVMS Data Collector nodes with the HOMERUNS password as part of their security triplets. The same is true of the Accounting Department's Data Analyzer, whose password is BATTERUP; it can access only OpenVMS Data Collector nodes with the BATTERUP password as part of their security triplets.

The Availability Manager sets a default password when you install the Data Analyzer. To change that password, you must use the OpenVMS Security Customization page (see Figure 7-21), which is explained in Chapter 7.

1.3.2 OpenVMS Data Collector Security

OpenVMS Data Collector nodes have the following security features:

• Availability Manager data-transfer security
  Each OpenVMS node running as a Data Collector has a file containing a list of security triplets. For Data Analyzer and Data Collector nodes to exchange data, the passwords on these nodes must match.
  In addition, the triplet specifies the type of access a Data Analyzer has. By specifying the hardware address of the Data Analyzer, the triplet can also restrict which Data Analyzer nodes are able to access the Data Collector.
  Section 1.3.3 explains security triplets and how to edit them.
• Availability Manager security log
  An OpenVMS Data Collector logs all access denials and executed write instructions to the operator communications manager (OPCOM). Messages are displayed on all terminals that have OPCOM enabled (with the REPLY/ENABLE command). OPCOM also puts messages in the SYS$MANAGER:OPERATOR.LOG file.
  Each security log entry contains the network address of the initiator. If access is denied, the log entry also indicates whether a read or write was attempted. If a write operation was performed, the log entry indicates the process identifier (PID) of the affected process.
• OpenVMS file protection and process privileges
  When the Availability Manager is installed, it creates a directory (SYS$COMMON:[AMDS$AM]) and sets directory and file protections on it so that only the SYSTEM account can read the files in that directory. For additional security on these system-level directories and files, you can create access control lists (ACLs) to restrict and set alarms on write access to the security files.
  For more information about creating ACLs, see the HP OpenVMS Guide to System Security.

To change security triplets on an OpenVMS Data Collector node, you must edit the AMDS$DRIVER_ACCESS.DAT file, which is installed on all Data Collector nodes. The following sections explain what a security triplet is, how the Data Collector uses it, and how to change it.

1.3.3.1 Understanding OpenVMS Security Triplets

A security triplet determines which nodes can access system data from an OpenVMS Data Collector node. The AMDS$DRIVER_ACCESS.DAT file on OpenVMS Data Collector nodes lists security triplets.

On OpenVMS Data Collector nodes, the AMDS$AM_CONFIG logical translates to the location of the default security file, AMDS$DRIVER_ACCESS.DAT. This file is installed on all OpenVMS Data Collector nodes.

A security triplet is a three-part record whose fields are separated by backslashes (\). A triplet consists of the following fields:

• A network address (hardware address or wildcard character)
• An 8-character alphanumeric password
  The password is not case sensitive (so the passwords "testtest" and "TESTTEST" are considered to be the same).
• A read, write, or control (R, W, or C) access verification code

The exclamation point (!) is a comment delimiter; any characters to the right of the comment delimiter are ignored.

Example

All Data Collector nodes in group FINANCE have the following AMDS$DRIVER_ACCESS.DAT file:

*\FINGROUP\R   ! Let anyone with FINGROUP password read 
               ! 
2.1\DEVGROUP\W ! Let only DECnet node 2.1 with 
               ! DEVGROUP password perform fixes (writes) 

On each Data Collector node on which you want to change security, you must edit the AMDS$DRIVER_ACCESS.DAT file. The data in the AMDS$DRIVER_ACCESS.DAT file is set up as follows:

      Network address\password\access 

Use a backslash character (\) to separate the three fields.

To edit the AMDS$DRIVER_ACCESS.DAT file, follow these steps:

1. Edit the network address.
   The network address can be either of the following:
   • Hardware address
     The hardware address field is the physical hardware address in the LAN device chip. It is used if you have multiple LAN devices or are running the HP DECnet-Plus for OpenVMS networking software on the system (not the HP DECnet Phase IV for OpenVMS networking software).
     For devices provided by HP, the hardware address is in the form 08-00-2B-xx-xx-xx, where the 08-00-2B portion is HP's valid range of LAN addresses as defined by the IEEE 802 standards, and the xx-xx-xx portion is chip specific.
     To determine the value of the hardware address on a node, use the OpenVMS System Dump Analyzer (SDA) as follows:

     $ ANALYZE/SYSTEM SDA> SHOW LAN

     
     These commands display a list of available devices. Choose the template device of the LAN device you will be using, and then enter the following command:

     SDA> SHOW LAN/DEVICE=xxA0

   • Wildcard address
     The wildcard character (*) allows any incoming triplet with a matching password field to access the Data Collector node. Use the wildcard character to allow read access and to run the console application from any node in your network.
     Caution: Use of the wildcard character for write-access or control-access security triplets enables any person using that node to perform system-altering fixes.
2. Edit the password field.
   The password field must be an 8-byte alphanumeric field. The Availability Manager forces upper-case on the password, so "aaaaaaaa" and "AAAAAAAA" are essentially the same password to the Data Collector.
   The password field gives you a second level of protection when you want to use the wildcard address denotation to allow multiple modes of access to your monitored system.
3. Enter R, W, or C as an access code:
   • R means READONLY access to the Data Analyzer.
   • W means READ/WRITE access to the Data Analyzer. (WRITE implies READ.)
   • C means CONTROL access to the Data Analyzer. CONTROL allows you to manipulate objects from which data are derived. (CONTROL implies both WRITE and READ.)

The following security triplets are all valid; an explanation follows the exclamation point (!).

*\1decamds\r   ! Anyone with password "1decamds" can monitor 
*\1decamds\w   ! Anyone with password "1decamds" can monitor or write 
2.1\1decamds\r ! Only node 2.1 with password "1decamds" can monitor 
2.1\1decamds\w ! Only node 2.1 with password "1decamds" can monitor and 
write 
08-00-2b-03-23-cd\1decamds\w ! Allows a particular hardware address to 
write 
08-00-2b-03-23-cd\1decamds\r ! Allows a particular hardware address to read 
node 

OpenVMS Data Collector nodes accept more than one password. Therefore, you might have several security triplets in an AMDS$DRIVER_ACCESS.DAT file for one Data Collector node. For example:

*\1DECAMDS\R 
*\KOINECLS\R 
*\KOINEFIX\W 
*\AVAILMAN\C 
 

In this example, Data Analyzer nodes with the passwords 1DECAMDS and KOINECLS are able to see the Data Collector data, but only the Data Analyzer node with the KOINEFIX password is able to write or change information, including performing fixes, on the Data Collector node. The Data Analyzer node with the AVAILMAN password is able to perform switched LAN fixes and other control functions.

You can choose to set up your AMDS$DRIVER_ACCESS.DAT file to allow anyone on the local LAN to read from your system, but to allow only certain nodes to write or change process or device characteristics on your system. For example:

*\1DECAMDS\R 
08-00-2B-03-23-CD\2NODEFIX\C 

In this example, any Data Analyzer node using the 1DECAMDS password can read data from your system. However, only the Data Analyzer node with the hardware address 08-00-2B-03-23-CD and the password 2NODEFIX can perform fixes and other control functions.

The Availability Manager performs these steps when using security triplets to ensure security among Data Analyzer and Data Collector nodes:

1. A multicast "Hello" message is broadcast at regular intervals to all nodes within the LAN indicating the availability of a Data Collector node to communicate with a Data Analyzer node.
2. The node running the Data Analyzer receives the message, returns a password to the Data Collector, and requests system data from the Data Collector.
3. The password and network address of the Data Analyzer are used to search the security triplets in the AMDS$DRIVER_ACCESS.DAT file.
   • If the Data Analyzer password and network address match one of the security triplets on the Data Collector, then the Data Collector and the Data Analyzer can exchange information.
   • If the Data Analyzer password and network address do not match any of the security triplets, then access is denied and a message is logged to OPCOM. (See Table 1-2 for more information on logging this type of message.) In addition, the Data Analyzer receives a message stating that access to that node is not permitted.

Table 1-1 describes how the Data Collector node interprets a security triplet match.

Sending Messages to OPCOM

The logical names shown in Table 1-2 control the sending of messages to OPCOM and are defined in the AMDS$LOGICALS.COM file on the Data Collector node.

To put these changes into effect, restart the Data Collector with the following command:

$ @SYS$STARTUP:AMDS$STARTUP RESTART

When the Data Analyzer detects problems on your system, it uses a combination of methods to bring these problems to the attention of the system manager. It examines both the types of data collected and how often it is collected and analyzed to determine problem areas to be signaled. Performance problems are also posted in the Event pane, which is in the lower portion of the System Overview window (Figure 1-1).

The following topics are related to the method of detecting problems and posting events:

• Collecting and analyzing data
• Posting events

This section explains how the Data Analyzer collects and analyzes data. It also defines related terms.

1.4.1.1 Events and Data Collection

The data that the Data Analyzer collects is grouped into data collections. These collections are composed of related data---for example, CPU data, memory data, and so on. Usually, the data items on the tabs (like the ones displayed in Figure 1-7) consist of one data collection.

Figure 1-7 Sample Node Summary

An event is a problem or potential problem associated with resource availability. Events are associated with various data collections. For example, the CPU Process data collection shown in Figure 1-8 is associated with the PRCCUR, PRCMWT, and PRCPWT events. (Appendix B describes events, and Appendix C describes the events that each type of data collection can signal.) For these events to be signalled, you must enable the CPU Process data collection, as described in Section 1.4.1.2.

Users can also customize criteria for events, which is described in Section 1.4.2.

1.4.1.2 Types of Data Collection

You can use the Data Analyzer to collect data either as a background activity or as a foreground activity.

Note that for either type of data collection, if you collect data for a specific node, only that node is affected. If you collect data for a group, all the nodes in that group are affected.

• Background data collection
  When you enable background collection of a specific type of data collection on a specific node, the Data Analyzer collects that data whether or not any windows are currently displaying data for that node.
  To enable background data collection, select the check box for a specific type of data collection on the Data Collection Customization page (Figure 1-8). Note that if the Customize window applies to all OpenVMS nodes, the data collection properties that you set are for all nodes. If the window applies to a specific node, the properties you set apply only to that node.
  Chapter 7 contains additional instructions for customizing data collection properties.

  Figure 1-8 Data Collection Customization

  ---

  
• Foreground data collection
  Foreground data collection occurs automatically when you open any data page for a specific node. To open a node data page, double-click a node name in the Node pane of the System Overview window (Figure 1-1). The Node Summary page is the first page displayed (by default); Figure 1-7 is an example. At the top of the page are tabs that you can select to display other data pages for that node.
  Foreground data collection for all data types begins automatically when any node data page is displayed. Data collection ends when all node data pages have been closed.
  Chapter 3 contains instructions for selecting nodes and displaying node data.

Data collection intervals, which are displayed on the Data Collection customization page (Figure 1-8), specify the frequency of data collection. Table 1-3 describes these intervals.

The Data Analyzer evaluates each data collection for events. The Data Analyzer posts events when data values in a data collection meet or exceed user-defined thresholds and occurrences. Values for thresholds and occurrences are displayed on Event Customization pages similar to the one shown in Figure 1-9. Thresholds and occurrences are described in the next section.

Figure 1-9 Sample Event Customization

Thresholds and occurrences are criteria that the Data Analyzer uses for posting events.

A threshold is a value against which data in a data collection is compared. An occurrence is a value that represents the number of consecutive data collections that meet or exceed the threshold.

Both thresholds and occurrences are customizable values that you can adjust according to the needs of your system. For details about how to change the values for thresholds and occurrences, see Chapter 7.

Relationship Between Thresholds and Occurrences

For a particular event, when the data collected meet or exceed the threshold, the data collection enters a threshold-exceeded state. When the number of consecutive data collections to enter this state meets or exceeds the value in the Occurrence box (see Figure 1-9), the Data Analyzer displays (posts) the event in the Event pane.

A closer look at Figure 1-9 shows the relationship between thresholds and occurrences. For the DSKERR, high disk device error count event, a threshold of 15 errors has been set. A value of 2 in the Occurrence box indicates that the number of errors during 2 consecutive data collections must meet or exceed the threshold of 15 for the DSKERR event to be posted.

This chapter provides the following information:

• How to configure and start the Availability Manager Data Collector
• How to start the Availability Manager Data Analyzer
• How to configure secure communications between the Data Analyzer and Data Server
• How to start the Availability Manager Data Server
• How to use the main System Overview window
• How to display basic node data
• How to get help when you need it
• How to print a Data Analyzer page

For information about installing the HP Availability Manager on OpenVMS or Windows systems, see the HP Availability Manager Installation Instructions. You can access these instructions from the documentation link at the Availability Manager web page at the following URL:

http://www.hp.com/products/openvms/availabilitymanager

The installation instructions also include an explanation of how to install and use both DECamds and the Availability Manager on the same system.

2.1 Configure and Start the Data Collector

Configuration tasks include defining logical names and protecting passwords. After you complete these tasks, you can start the Data Collector. The following sections describe all of these operations.

2.1.1 Defining Logical Names

OpenVMS kits for DECamds Version 7.3-2B and Availability Manager Version 3.1 provide a template file that system managers can modify to define the logical names used by the Data Collector. You can copy the file SYS$MANAGER:AMDS$SYSTARTUP.TEMPLATE to SYS$MANAGER:AMDS$SYSTARTUP.COM and edit it to change the default logicals that are used to start the Data Collector and to find its configuration files.

The most common logicals, especially in a mixed-environment cluster configuration, are the ones shown in Table 2-1:

The Data Collector on a node transmits multicast "Hello" messages for any Data Analyzer or Data Server on the extended LAN to receive. The rate at which these messages are transmitted is regulated by the settings of the following logicals:

AMDS$RM_DEFAULT_INTERVAL
AMDS$RM_SECONDARY_INTERVAL

These logicals are contained in the file SYS$MANAGER:AMDS$LOGICALS.COM. The shorter the time interval, the faster the node is discovered and configured.

2.1.2 Protecting Passwords

To change passwords to allow a Data Analyzer to monitor a node, edit the following file:

SYS$MANAGER:AMDS$DRIVER_ACCESS.DAT 

The passwords section of the file is close to the end of the file, after the Password documentation section. The passwords in this file correspond to the passwords in the Security page shown in Section 7.9.1. Note that you can specify a list of passwords in this file. See the comments in the file for details.

2.1.3 Starting the Data Collector

Beginning with OpenVMS Version 7.2, the files needed to run the Data Collector on OpenVMS nodes are shipped with the OpenVMS operating system. However, if you want the latest Data Collector software, you must install it from the Availability Manager Data Collector kit. Once the Data Collector is running on a node, you can monitor that node using DECamds or the Availability Manager Data Analyzer.

For the Data Collector to access requests to collect data and to support the Data Analyzer, start the Data Collector by entering the following command:

     $ @SYS$STARTUP:AMDS$STARTUP START 

To start the Data Collector when the system boots, add the following command to the SYS$MANAGER:SYSTARTUP_VMS.COM file:

     $ @SYS$STARTUP:AMDS$STARTUP START 

If you make changes to either the AMDS$DRIVER_ACCESS.DAT or AMDS$LOGICALS.COM, restart the driver to load the changes. Enter the following command:

     $ @SYS$STARTUP:AMDS$STARTUP RESTART 

$ RUN SYS$SYSTEM:SYSMAN 
SYSMAN> SET ENVIRONMENT/CLUSTER 
SYSMAN> DO @SYS$STARTUP:AMDS$STARTUP START 
SYSMAN> EXIT 
$ 

This section describes what you need to do after the Availability Manager Data Analyzer is installed. Starting the Data Analyzer is somewhat different on OpenVMS than on Windows systems. However, on both systems, starting the Data Analyzer automatically starts the Java graphical user interface (GUI), which allows you to view information that is collected from Data Collectors running on OpenVMS nodes.

The following sections describe the sequence of steps required to start the Data Analyzer on an OpenVMS node and on a Windows node.

2.2.1 Starting the Data Analyzer on an OpenVMS Node

To start a Data Analyzer on an OpenVMS node, make sure that:

• The Data Analyzer is installed on the node from which you want to monitor other nodes.
• The Data Collector is started (see Section 2.1.3).

Starting the Data Collector accomplishes the following important tasks:

• Defines the various AMDS$* logicals needed by the Data Analyzer.
• Allows the Data Analyzer to communicate with the Data Collector on the network.

To start the Data Analyzer, enter the following command:

   $ AVAIL/ANALYZER 

The Data Analyzer displays the Network Connection dialog box, which is shown in Figure 2-20.

To start the Data Analyzer on a Windows node, first make sure that the Availability Manager Windows kit is installed on the node.

To start the Data Analyzer, follow these steps:

1. Click
   Start ---> Programs.
2. Select Availability Manager.
3. Select Data Analyzer Startup.

The Data Analyzer displays the Network Connection Dialog box, which is shown in Figure 2-20.

2.3 Do You Need to Set Up a Data Server?

At this point, you must determine whether you need to use a Data Server to communicate with the Data Collectors. For an overview of what a Data Server is and how it works, see Section 1.2.2.

If the analyzer system is on the same LAN as the Data Collectors, you can use a network adapter on the analyzer system to connect with the Data Collectors. If this is the case, you do not need to set up the Data Server. To continue starting the Data Analyzer without a Data Server, go to Section 2.6.

If the Data Analyzer is on a different LAN than the Data Collectors, you must set up the Data Server on a server system that is on the same LAN as the Data Collectors. To set up secure communication between the Data Analyzer and Data Server, see Section 2.4.

To collect data over a WAN, the Data Analyzer communicates with a Data Server. The Data Server is a Java-based program that runs on OpenVMS or Windows. Except for differences in starting the Data Server on OpenVMS and Windows, the following section applies to both operating systems.

The Availability Manager uses an encrypted connection for secure communication between the Data Analyzer and the Data Server. The following sections describe how to set up the Data Analyzer and Data Server to use a secure communication link.

2.4.1 Introduction to Secure Communications

The Availability Manager uses Transport Layer Security (TLS) Version 1 for secure communication between the Data Analyzer and the Data Server. TLS is an extension of Secure Sockets Layer (SSL) Version 3.0, which is the most widely used protocol for security on the web.

TLS uses public key cryptography (also called asymmetric cryptography) to guarantee secure communication over a network. This type of cryptography uses an encryption algorithm that produces a pair of keys:

• A public key provides authentication, and is made public to any interested party as a trusted certificate.
• A private key that works with trusted certificates to provide privacy and data integrity

What one key encrypts, only the other key can decrypt. Together, these two keys are known as an asymmetric key pair.

Key Pairs, Key Stores, and Trust Stores

Before you can use the Data Server, you must create an asymmetric key pair. This key pair is associated with the Data Server, and is used by the Data Server and Data Analyzer to establish an encrypted communication link between them.

The Data Server stores the public and private key associated with it in a key store . The Data Server key store is the file AM$KeyStore.jks and resides on the server system. On OpenVMS systems, this file is in the AMDS$AM_CONFIG: directory. On Windows systems, the key store is in the installation folder. Currently, HP supports configurations in which the Data Server has only one key pair in a key store.

The Data Server public key is also stored by the Data Analyzer in a trust store on the analyzer system. The Data Analyzer trust store is the file AM$TrustStore.jks. On OpenVMS systems, this file is in the AMDS$AM_CONFIG: directory. On Windows systems, the trust store is in the installation folder. A trust store for a particular Data Analyzer holds the public key for each Data Server with which it communicates.

You create and store the key pair after installing either the combined kit (for OpenVMS) or the Availability Manager kit (for Windows). The next sections describe how to perform the following tasks:

• Creating the key pair from either the server or analyzer system
• Store the key pair in a key store on a server system
• Store the public key in a trust store on an analyzer system

The key store and trust store are created and maintained by dialog boxes in the Data Analyzer. The Data Analyzer is used for key management because it is the part of the Availability Manager that uses a GUI interface. By using the GUI interface, keys are managed the same way on OpenVMS and Windows platforms. This also keeps the Data Server from having the overhead of the dialog boxes used for creating and maintaining key and trust stores.

There are two basic methods of setting up secure communications. Both methods create a key store for the Data Server and a trust store for the Data Analyzer. The difference is that one creates the key store using the server system, and the other creates the key store from the analyzer system. Using one method or the other is sufficient to set up secure communications between the Data Analyzer and Data Server.

2.4.2.1 Setup Using the Server System

Creating the key store from the server system is the simplest method. You create the key store and export the public key using the Data Analyzer on the server system, copy the public key to the analyzer system, and import the public key with the Data Analyzer on the analyzer system. For a description of this method, see Section 2.4.3.

Using this method assumes that you can use the Data Analyzer's GUI interface on the server system. You can start the Data Analyzer on the server system and display the GUI on the following:

• the server graphics console
• another OpenVMS system that does have a graphics console
• a Windows system that has software to accept and display an X Windows GUI

If this is not possible, use the alternate method to create and maintain key stores in Section 2.4.2.2.

2.4.2.2 Setup Using the Analyzer System

With this method, you create the key store and export the public key using the Data Analyzer on the analyzer system, and copy the key store to the server system. This method is described in Section 2.4.4.

2.4.3 Steps for Setting Up Secure Communications from the Server System

The following section describes how to set up the Data Server from the server system. It also describes the key setup for the Data Analyzer that runs on the server system. The procedure involves the following tasks:

• Creating the key pair for the Data Server, including the option of generating and storing the trust store for the Data Analyzer on the server system,
• Storing the key pair in the Data Server's key store on the server system
• Storing the public key for another Data Analyzer to use

When you complete these steps, the Data Server can accept connections from any Data Analyzer on the server system or on other systems.

2.4.3.1 Creating the Key Pair for the Data Server

1. Start the Data Analyzer on the server system according to the instructions in Section 2.2. When the Data Analyzer starts, it displays the Network Connection dialog box as shown in Figure 2-1.

   Figure 2-1 Network Connection Dialog Box

   ---

   
   
2. From the Server menu, select Key Store... to open the default key store for this system.
   The Availability Manager displays the Key Store Management dialog box as shown in Figure 2-2.

   Figure 2-2 Key Store Management Dialog Box

   ---

   
   
3. In the Key Store Management dialog box, click New Key... to display the Generate New Key Pair dialog box as shown in Figure 2-3.

   Figure 2-3 Generate New Key Pair Dialog Box

   ---

   
   To create a new key pair, fill in the fields in this dialog box.
   The information you enter in the Generate New Key Pair dialog box includes fields that pertains to an X.500 Distinguished Name. HP recommends that you enter the name of the server system in the Server Name field (CN) and in Alias field. ("Alias" is simply a name that is used to track items in the key store and is not part of the generated key.)
   Currently, the Availability Manager does not verify whether or not a key has expired. Therefore, the Validity field is not used. However, for the field to work in future versions, HP recommends that you enter a large value if you are creating a key that must be valid for a long time.
   To run the Data Analyzer on the server system and have it connect to the Data Server on the server system, check the Default Trust Store check box. This creates a trust store for the Data Analyzer that contains the public key for accessing the Data Server on the server system.
   When you finish entering information to create a new key pair for the Data Server, click Add (it might take a few seconds to create the key). If you checked the Default Trust Store check box, the default trust store for this key pair is created for the Data Analyzer running on the server system.
   The Key Store Management dialog box shown in Figure 2-4 now displays one key pair, reflecting the information you entered in the Generate New Key Pair dialog box.

   Figure 2-4 Key Store Management Dialog Box Showing Key Pair

   ---

   
   If the only system you want to run the Data Analyzer is the server system, then do the following:
   1. Click on OK in the Key Store Management dialog box to save the key store on the server system.
   2. Follow the instructions in Section 2.6 to start and configure the Data Analyzer.

To run the Data Analyzer on other systems, see Section 2.4.3.2.

2.4.3.2 Export the Public Key for Other Data Analyzers

To run the Data Analyzer on other systems, and to connect to the Data Server on this system, you must export the public key for the Data Server as a trusted certificate. To do this, click the key pair name in the Key Store Management dialog box. This action enables the Export... button. Click Export... to export the public key in a trusted certificate. The Availability Manager displays the Export Certificate dialog box as shown in Figure 2-5.

Figure 2-5 Export Certificate Dialog Box

Store the trusted certificate in the folder and file name of your choice. Any file name with a CER extension works, although naming the file the same as the server alias can make it easier to identify. Click Export to complete this process.

To save the key store on the server system, click OK in the Key Store Management dialog box. Then see Section 2.4.5 to import the trusted certificate into the Data Analyzer trust store.

2.4.4 Steps for Setting Up Secure Communications from the Analyzer System

The process for setting up the Data Server from an analyzer system involves the following tasks:

• Creating the key store for the Data Server on the server system.
• Exporting the public key as a trusted certificate for other analyzer systems.
• Saving the key store.
• Copying the key store to the server system.
• Delete the key and trust store from the analyzer system.
• Exporting the public key to the server system from an existing server system using an analyzer system.

Start the Data Analyzer on the analyzer system. When the Data Analyzer starts, it displays the Network Connection dialog box as shown in Figure 2-6.

Figure 2-6 Network Connection Dialog Box

From the Key Stores menu, click New Trust or Key Store.... The Availability Manager displays the Key Store Management dialog box, shown in Figure 2-7.

Figure 2-7 Key Store Management Dialog Box

In the Key Store Management dialog box, click New Key... to display the Generate New Key Pair dialog box as shown in Figure 2-8. To create a new key pair, fill in the fields in this dialog box. For a description of these fields, see Section 2.4.3.1.

Figure 2-8 Generate New Key Pair Dialog Box

When you finish entering information in the Generate New Key Pair dialog box, click Add (it might take a few seconds to create the key). If you checked the Default Trust Store check box, the default Trust Store for this key pair is created for the Data Analyzer running on the this analyzer system.

The Key Store Management dialog box (Figure 2-9) now displays the new key pair, reflecting the information you entered.

Figure 2-9 Key Store Management Dialog Box with One Entry

This step finishes the setup needed for this analyzer system. If this is the only Data Analyzer that needs to connect to this Data Server, go to Section 2.4.4.4.

2.4.4.2 Exporting the Public Key for Analyzer Systems

For other Data Analyzers that need to connect to the Data Server, export the public key as described in this section.

In the Key Store Management dialog box, select the Data Server key pair by clicking the key entry. This enables the Export... button in the dialog box. Click Export... to extract the Data Server's public key and store it in a file as a trusted certificate.

The Export Certificate dialog box is displayed as shown in Figure 2-10.

Figure 2-10 Export Certificate Dialog Box

Store the trusted certificate in the folder and file name of your choice. Any file name with the CER extension works, although accepting the default can make the file easier to identify. Click on the Export button to complete this process.

Now that you have created the key pair for the Data Server, you must save the pair in a key store. In the Key Store Management dialog box, select the Key Store menu, and then select Save. This displays the Save Key Store dialog box as shown in Figure 2-11.

Figure 2-11 Save Key Store Dialog Box

Save the key store in the folder and file name of your choice. Any file name with a JKS extension works, although naming the file the same as the server alias can make the file easier to identify. Enter this file name in the File Name: field, and click Save to save the key store. In the Key Store Management dialog box, click Cancel to dismiss the dialog box.

2.4.4.4 Copying the Key Store to the Server System

The key store is now ready for the server system. Copy the file to the server system. If you use FTP to transfer the file, be sure to use the binary transfer mode.

Once the file is copied, move it to the location and file name that the Data Server looks for when it starts. On OpenVMS, the location is AMDS$AM_CONFIG: directory. On Windows, the location is the installation directory. Make sure that the file is named AM$KeyStore.jks.

2.4.4.5 Delete the key and trust store from the analyzer system

Once you have created the key store and copied it to the server system, it is recommended that you delete the key and trust store on the analyzer system. This sets up the analyzer system to create a key store for another Data Server, or to create the trust store by importing the trusted certificates from each Data Server into the Data Analyzer.

This concludes the Data Server setup on the server system. If you want to create a key store for another Data Server, go to Section 2.4.4. Otherwise, go to Section 2.4.5, which describes how to import the Data Server's public key into the trust store of other Data Analyzers.

The next section describes how to obtain the public key from an existing Data Server. This step allows the Data Analyzer to connect to the Data Server.

2.4.4.6 Obtaining the Public Key from an Existing Data Server

This section describes how to obtain a Data Server's public key from the analyzer system.

2.4.4.6.1 Copy the Key Store from the Server System

This step extracts the Data Server public key from the key store by exporting it to a trusted certificate.

Start the Data Analyzer on the analyzer system. When the Availability Manager starts, it displays the Network Connection dialog box as shown in Figure 2-12.

Figure 2-12 Network Connection Dialog Box

From the Key Stores menu, select Open Trust or Key Store... to open the Open Key or Trust Store dialog box as shown in Figure 2-13.

Figure 2-13 Open Key or Trust Store Dialog Box

In this dialog box, locate the key store file by selecting the name of the key store file, and clicking Open. The opened key store is displayed in the Key Store Management dialog box as shown in Figure 2-14.

Figure 2-14 Key Store Management Dialog Box

Select the key pair entry in the dialog box. This enables the Export... button. Click Export... to export the public key of the key pair into a trusted certificate. The Availability Manager displays the Export Certificate dialog box as shown in Figure 2-15.

Figure 2-15 Export Certificate Dialog Box

Store the trusted certificate in the folder and file name of your choice. Any file with the CER extension works, although accepting the default can make the file easier to identify. Click Export to complete this process. You now have the trusted certificate.

This section describes how to set up a trust store for a Data Analyzer to connect to an existing Data Server. The steps involve the following tasks:

• Obtaining the Data Server's public key from its key store as a trusted certificate.
• Copying the trusted certificate to the analyzer system.
• Importing the trusted certificate into the Data Analyzer's trust store.

First enter the Data Server's public key into the trust store of the Data Analyzer. This transfer involves exporting the key into a trusted certificate from the key store, and importing the key into the Data Analyzer's trust store.

The following sections describe how to export the the public key into a trusted certificate. If you need to export the public key, determine which of the following applies to you.

• Section 2.4.3.2, Export the Public Key for Other Data Analyzers
• Section 2.4.4.2, Exporting the Public Key for Analyzer Systems
• Section 2.4.4.6.2, Export the Key Store Public Key to a Trusted Certificate

Make sure you have the Data Server's public key in a trusted certificate for the next step.

2.4.5.2 Copying the Trusted Certificate

Copy the trusted certificate from the server system to the analyzer system. Note that the trusted certificate contains binary data, so you must use binary mode if FTP is the file transport. The certificate is now ready for importing to the Data Analyzer's trust store.

2.4.5.3 Importing the Data Server Public Key

Start the Data Analyzer on the analyzer system. From the Analyzer menu, select Trust Store to open the default trust store for this system. The Availability Manager displays the Trust Store Management dialog box as shown in Figure 2-16.

Figure 2-16 Trust Store Management Dialog Box

Click Import... to import the trusted certificate. The Availability Manager displays the Import Certificate dialog box as shown in Figure 2-17.

Figure 2-17 Import Certificate Dialog Box

Select the name of the trusted certificate, and click Import. The Availability Manager displays the Assign Alias for Certificate dialog box as shown in Figure 2-18.

Figure 2-18 Assign Alias for Certificate Dialog Box

This dialog box displays the trusted certificate. Enter the alias name for the certificate in the Assign Alias field. Although you can put any text in this field, it is best to choose the same alias name that the Data Server uses. Then click OK to continue. The Availability Manager displays the Trust Store Management dialog box with the imported key as shown in Figure 2-19.

Figure 2-19 Trust Store Management Dialog Box

In the Trust Store Management dialog box, click OK to save the trusted certificate in the Data Analyzer trust store.

This sets up the Data Analyzer to connect to a Data Server. The Data Analyzer supports connections to multiple Data Servers. To connect to multiple Data Servers, export the public key for each Data Server and import it into the Data Analyzer.

This completes the Data Analyzer key configuration. You are now ready to run the Data Analyzer and connect to the Data Server.

2.5 Starting the Data Server

This section describes tasks you must perform after the Availability Manager Data Server is installed. Starting the Data Server is somewhat different on OpenVMS than on Windows systems. However, on both systems, the Data Server listens for connections from Data Analyzers once it is started.

The Data Server is designed to run in a minimal environment. It only outputs text messages to log various events and Data Analyzer connections. Because of this design, it can be run in a batch job on OpenVMS, or as a startup task on Windows.

The following sections contain the sequence of steps required to start the Data Server on an OpenVMS node and a Windows node.

The first step is to decide which platform is to run the Data Server: Windows or OpenVMS.

2.5.1 Starting the Data Server on an OpenVMS System

To start a Data Server on an OpenVMS System (Alpha or I64), make sure the following conditions are met:

• The Data Server is installed on a node that is on the same LAN as your OpenVMS systems.
• The Data Collector is started (see Section 2.1.3).

Starting the Data Collector is important for these reasons:

• Defines the various AMDS$* logicals needed by the Data Server.
• Allows the Data Server to communicate to the Data Collector on the network.

After you install and configure the Data Collector and Data Server and start the Data Collector, enter the following command to start the Data Server:

   $ AVAIL/SERVER 

To install and configure the Availability Manager, follow the steps in the HP Availability Manager Installation Instructions. Then, to start the Data Server, click Click Start -> Programs -> HP Availability Manager -> Data Server Startup.

2.5.3 Data Server Port and Firewalls

If you are running a firewall on your server system, ensure that the firewall allows communication over the port the Data Server uses. The default port number is 9819, and the type of connection for the port is TCP.

2.6 Using the Network Connection Dialog Box to Start Collecting Data

The following section describes the steps needed to get the Data Analyzer to connect to one or more network adapters, or connect to one or more Data Servers. The Data Analyzer supports any combination of available network adapters and Data Servers.

These steps assume that the Data Servers are already running on the server systems.

Start the Data Analyzer on the analyzer system as described in Section 2.2. The Availability Manager displays the Network Connection dialog box, shown in Figure 2-20.

Figure 2-20 Network Connection Dialog Box

Figure 2-20 shows two entries for the two network adapters on this particular system. The last entry is where you enter the IP address and port number of a Data Server. To use one or more of these network adapters, check the check box to the left of each network adapter, and click OK. The Data Analyzer starts, using the network adapters you have chosen. To start using the Data Analyzer, see the instructions in Section 2.8.

To connect to one or more Data Servers, enter the IP address of each server, along with the IP port that the Data Server uses for communication. There are a number of possible forms for the IP address:

• Alphanumeric IP address - Alpha1.denver.newscorp.com
• Numeric IP address - 136.132.15.32
• WINS entry for a Windows system - WXPSRV1
• Analyzer system name synonym - Localhost

The default IP address shown in the dialog box is "localhost". Localhost is a synonym for the IP address of the Analyzer system itself. Use the "localhost" default or enter the IP address of the Data Server, the IP port the Data Server is using in the Port: field, and click on the plus sign button to register the entry. The data for the new Data Server entry is displayed in the dialog box. You can repeat this process to enter all the Data Servers you want to use.

Figure 2-21 shows an example of this procedure. The IP address entered is Aslan, the WINS entry for the Data Server system, and the port number entered is 9819.

Figure 2-21 Network Connection Dialog Box with One Data Server Entry

Figure 2-22 shows the result of adding a second Data Server using the numeric form of the IP address.

Figure 2-22 Network Connection Dialog Box with Two Data Server Entries

Figure 2-23 shows the result of adding a third Data Server using the alphanumeric form of the IP address.

Figure 2-23 Network Connection Dialog Box with Three Data Server Entries

To remove a Data Server entry from the Network Connection dialog box, click the delete button (X) to the right side of the Data Server entry.

To start collecting data, check the network adapter and Data Server entries you want to use, and click OK. This process is described in Section 2.7.

2.6.1 Additional Information About Key Stores

This section contains some additional information about handling keys, key stores and trust stores.

2.6.1.1 Clarification of Network Connection dialog box Menus

Note the following:

• The Key Store menu item on the Server and the Key Stores menu open the default Data Server key store (AM$KeyStore.jks). This default key store name is what the Data Server uses when it starts. You can save key stores with other file names, but when you copy the key store to the server system for the Data Server to use, you must rename it to the default key store name.
• The Trust Store menu item on the Analyzer and Key Stores menus and the Trust Store button open the default Data Analyzer trust store (AM$TrustStore.jks). This default trust store name is what the Data Analyzer uses when it starts. You can save trust stores with other file names, but when you copy the trust store to the analyzer system for the Data Analyzer to use, you must rename it to the default trust store name.
• The other menu items on the Key Stores menu open generic key or trust stores that you are prompted to name when you open or save any of them.

The Availability Manager allows you to open multiple key and trust stores using the menus on the Network Connection dialog box. The Key Store and Trust Store Management dialog boxes allow you to drag and drop items interchangeably between dialog boxes (and to the file system or desktop on Windows). This operation can make import and export easier if you open the key and trust stores locally or if you use network shares to open them.

2.6.1.3 Certificates

The certificate that you create is a "self-signed" one. This means that the person who creates the certificate also signs off on its legitimacy. This type of certificate is also called a root certificate.

2.7 Choosing Network Connections for Collecting Data

When you start the Data Analyzer, it displays the Network Connection dialog box. This dialog box shows the available network adapters on the system, and any Data Servers that have been entered. You can choose which networks adapters and Data Servers the Data Analyzer uses for collecting data by check the check box of each entry.

Figure 2-24 shows a Network Connection dialog box with the two available network adapters on the system, and three Data Servers. Three of the entries are checked. Section 2.8 uses this example to document how to use the Data Analyzer.

Figure 2-24 Sample Network Connection Dialog Box with Three Checked Entries

After you click OK on the Network Connection dialog box, the Data Analyzer displays the System Overview window Figure 2-25 and monitors the network for multicast "Hello" messages from nodes running the Data Collector. It follows these steps:

1. After receiving a multicast "Hello" message from the Data Collector, the Data Analyzer attempts to connect to a node. This is called the attempting collection state.
   The Data Analyzer notifies you of this and other states in the System Overview window, which is shown in Figure 2-25.
2. The Data Collector performs a security check on the Data Analyzer connection attempt.
   • If the Data Analyzer passes the security check while the Availability Manager is attempting the connection, the connection succeeds, and data collection starts. This is called the data collection state.
   • If the Data Analyzer fails the security check, the node is in the connection failed state.
3. While the Data Analyzer collects data, if a node goes down, or a network connection fails between the graphical user interface and the node, that node is placed in the path lost state.

The colors of the icons preceding each node name in Figure 2-25 indicate the state of the node.

Figure 2-25 System Overview Window

The color code of each node state is explained in Table 2-2.

The System Overview window is divided into two segments, or panes: the Group/Node pane and the Event pane.

2.8.1 Using the Group/Node Pane

When you start the Data Analyzer, the System Overview window (see Figure 2-25), displays information on connection lines at the top of the pane (that is, lines starting with "Device", "Aslan" and "16.212.8.229" in Figure 2-25). The items on these lines measure throughput and congestion on each connection. The following table describes the column headings.

If the number of packets waiting or discarded is consistently large, you might notice that the data displayed in the application updates at a slower rate. In extreme cases, nodes might turn black, indicating a lost connection with the node when, in reality, the problem is the congestion between the Data Analyzer and the Data Server.

If you have a problem with congestion, consider scaling back the number of nodes or the amount of data being collected, or lengthening the collection intervals.

The rest of the Group/Node pane displays information about the OpenVMS groups and nodes that the Data Analyzer has found. By default, within each group, the Data Analyzer displays the nodes with which it can establish a connection. (If the Data Analyzer finds Windows nodes, those are also displayed.)

2.8.1.1 Setting Up Groups

Groups are set up during installation on Data Collector nodes and are user-definable. Be sure to define groups by cluster membership. If a node is not a member of a cluster, then you can define a group by function, type of hardware, or geographical location.

If you want to change the groups being monitored, you need to use a customization option to make changes. See Section 7.4.1 for instructions.

Groups---and the nodes in each group with which the Data Analyzer is able to establish a connection---are displayed in the Group/Node pane of the System Overview window (see Figure 2-25).

To display only groups in the Group/Node pane, click the handle in front of a group name to a horizontal position, and the nodes in that group are removed, as shown for both groups in Figure 2-26. (Clicking the handle into a vertical position displays nodes again.)

Figure 2-26 Group Overview Pane

The numbers in parentheses after "OpenVMS" (in the Group/Node pane of the System Overview window) are the following:

• The first number in parentheses is the total number of groups that are listed.
• The second number in parentheses is the total number of nodes in all the listed groups with which the Data Analyzer can establish a connection.

On each group name row, following the name of the group, the number in parentheses is the number of nodes in that group with which the Data Analyzer has established a connection.

On a group name row under the OS Version heading are color-coded numbers indicating the number of nodes in that group that are one of five color-coded states. These states are explained in Table 2-2.

Additional summary information about the entire group is on the group line. CPU, MEM, BIO, and DIO numbers are averages. The rest of the number are totals for all of the nodes in the group.

Notice the small triangle in the BIO heading in Figure 2-26. The direction of the triangle indicates that the nodes are sorted in descending order of BIO rates. Click on the triangle to reserve the sort order, or click on another column header to select a new item on which to sort data.

In the Group/Node pane, only nodes within a group are sorted. The groups remain in alphabetical order. You can sort groups in the Group Overview window by changing the sort order of one of the data column headings (see Figure 2-26).

2.8.2 Displaying Node Information

The Group/Node pane of the System Overview window allows you to focus on resource usage activity at a high level and to display more specific data whenever you want. This section explains the basic use of the Group/Node pane. For more information, see Chapter 3.

2.8.2.1 Displaying Summary Node Information

Even when nodes are not displayed on the System Overview window or the Group/Node pane, you can display important node information by placing the cursor over a group name or icon. By holding the cursor over the KOINE group name, for example, the tooltip similar to the one shown in Figure 2-27 is displayed, containing summary node information.

Figure 2-27 Tooltip Example: Summary Node Information

Possible tooltip colors and their meanings are in Table 2-3.

The Group/Node pane is designed to display monitored nodes in a single pane. This format works well for sites that have relatively few nodes to monitor. However, for large sites that have many groups and nodes, scrolling through the display can be time-consuming. To help those with large sites, two additional windows are available:

• The Group Overview window
• The Single-Group window

The first window to help you view large sites is the Group Overview window. To view all the group name row data easily, click on the View menu at the top of the page and select "Group Overview." The Group Overview window that is displayed (Figure 2-28) is similar to the Group Overview pane in Figure 2-26.

Figure 2-28 Group Overview Window

This display is designed to provide an overview of all the groups being monitored. If you want more information about a group, place the cursor over the group name or icon. A tooltip is displayed with additional information about nodes in the group similar to the one displayed in Figure 2-27.

You can also double-click a group name to display a Single-Group window, as explained in Section 2.8.2.3.

2.8.2.3 Displaying a Single-Group Window

The second window to help you view large sites is the Single-Group window. This display shows the nodes in one group (see Figure 2-29).

To obtain this display, you can also right-click the group name in the Group/Node pane and select the "Display" option. A separate window appears with only the nodes in the group you have selected (see Figure 2-29). This window is useful in simultaneously displaying groups that are not adjacent in the list in the Group/Node pane.

Figure 2-29 OpenVMS Single-Group Window

Within each group of nodes displayed, the Data Analyzer displays all the nodes with which it can communicate. If some nodes in the group are not displayed, it is because the Data Analyzer has not received a multicast "Hello" message from the Data Collector on that node.

The display includes the following items:

• A list of the nodes in the group along with summary data for each node. In Figure 2-25, the Debug cluster group contains 9 nodes.
• A color-coded monitor icon preceding each node name indicates the state of the node. See Table 2-2 for explanations of states these colors indicate.
• For various node data items, some graphs indicate the percentage of an item that is being used; other graphs are totals.
  Green graphs indicate percentages below a customized threshold; red graphs indicate percentages above a customized threshold. Some data items are numbers, not percentages; for example, CPUs, CPU queues, and events.

More information about node data is in Chapter 3.

Somewhat different information is displayed for a group of Windows nodes. For more information, see Section 3.1.2.

2.8.2.4 Focusing On a Specific Node

To display more information about an individual node, double-click a node name or in the Single-Group window or the Group/Node pane. You can also right-click a node name and select the "Display..." option. The Data Analyzer displays the Node Summary page shown in Figure 2-30. (The data on this page is explained in more detail in Chapter 3.)

Figure 2-30 OpenVMS Node Summary

At the top of the Node Summary page are tabs that correspond to types of node data displayed in the Group/Node pane. If you double-click a field under a column heading in the Group/Node pane, the Data Analyzer displays a page that provides more information about that field. For example, if you click a value under "CPU", the Data Analyzer displays a page similar to the one shown in Figure 3-6.

2.8.2.5 Specifying Data to Be Collected

By default, the only data collected for a node is the data displayed in the Node pane (Figure 2-29). This data is called a node summary data collection. The events in the Event pane of the System Overview window (see Figure 2-25) are produced when node summary data is processed. See Appendix C for a list of events associated with node summary data.

If you want to signal additional events that are listed in Appendix C, you must collect the data associated with those events. To collect this data by default, you must enable background data collection for the data. Background and foreground data collections are explained in more detail in Section 1.4.1.2.

For OpenVMS nodes, if you want background data collection (and the associated event detection), you must turn on data collection for each type of data you want to collect. On Windows nodes, background data collection is always enabled and cannot be turned off.

To turn on various types of data to be collected, follow these steps:

1. In the System Overview window (Figure 2-25), click the Customize menu.
2. Click Customize OpenVMS....
3. Click the Data Collection tab.

The Data Analyzer then displays the Data Collection Customization page (Figure 2-31).

Figure 2-31 Data Collection Customization

The following types of data are collected by default:

• Node summary
• Single disk
• Single process

To turn on a type of data collection, select the checkbox for that type of data collection in the "Collect" column. For example, to collect CPU process data, check the checkbox for "CPU process" in the Collect column. Clicking the checkbox again clears it.

When you click a data collection name, the Explanation section at the bottom of the page tells where the data for a particular data collection is displayed. Table 7-3 summarizes this information.

You cannot turn off the collection of single disk and single process data. These types of data are collected by default when you open a Single Disk Summary page or a Process Information page, respectively.

On the Data Collection Customization page, you can change the intervals at which data is collected. Collection intervals are explained in Chapter 7.

2.8.2.6 Sorting Data

You can sort data in many OpenVMS displays. The following list provides some examples. To sort the values in a field, click the corresponding column heading. To reverse the sort order, click the column heading again.

• Event pane of the System Overview window (Figure 2-25)
• CPU Process Summary pane (Figure 3-8)
• Memory page (Figure 3-10)
• Bottom pane of I/O Summary page (Figure 3-12)
• Disk Status Summary page (Figure 3-14)
• Disk Volume Summary page (Figure 3-16)

Depending on the field, you can sort data alphabetically or numerically. An alphabetical sort is performed using ASCII character values; for example, dollar signs ($) precede letters in the sort order.