HP OpenVMS Availability Manager User's Guide

7.8.1 Customizing Events

You can change the values for any data that is available---that is, not dimmed---on this page. The following table describes the data you can change:

Data Description

Severity Controls the severity level at which events are displayed in the Event pane of the System Overview window (Figure 2-1). By default, all events are displayed. Increasing this value reduces the number of event messages in the Event pane of the System Overview window (Figure 2-1) and can improve perceived response time.

Occurrence Each Availability Manager event is assigned an occurrence value, that is, the number of consecutive data samples that must exceed the event threshold before the event is signaled. By default, events have low occurrence values. However, you might find that a certain event indicates a problem only when it occurs repeatedly over an extended period of time. You can change the occurrence value assigned to that event so that the Availability Manager signals the event only when necessary.
For example, suppose page fault spikes are common in your environment, and the Availability Manager frequently signals intermittent HITTLP, total page fault rate is high events. You could change the event's occurrence value to 3, so that the total page fault rate must exceed the threshold for three consecutive collection intervals before being signaled to the event log.
To avoid displaying insignificant events, you can customize an event so that the Availability Manager signals it only when it occurs continuously.

Threshold Most events are checked against only one threshold; however, some events have dual thresholds, where the event is triggered if either one is true. For example, for the LOVLSP, node disk volume free space is low event, the Availability Manager checks both of the following thresholds:

Number of blocks remaining
Percentage of total blocks remaining

Escalation actions You can enter one or more of the following values:

User: If the event occurs, the Availability Manager refers to the User Action field to determine what action to take.
OPCOM: If the event occurs, and certain conditions are met (see Section 7.7), the Availability Manager passes that event to OPCOM. (Data Analyzer on OpenVMS only)
HP OpenView: If the event occurs and certain conditions are met (see Section 7.7), the Availability Manager passes that event to HP OpenView. (OpenView agents must be installed and configured on the Data Analyzer node.)

User Action When the Event escalation action field is set to User, User Action is no longer dimmed. You can enter the name of a procedure to be executed if the event displayed at the top of the page occurs. To use this field, see the instructions in Section 7.8.2.

Data	Description
Severity	Controls the severity level at which events are displayed in the Event pane of the System Overview window (Figure 2-1). By default, all events are displayed. Increasing this value reduces the number of event messages in the Event pane of the System Overview window (Figure 2-1) and can improve perceived response time.
Occurrence	Each Availability Manager event is assigned an occurrence value, that is, the number of consecutive data samples that must exceed the event threshold before the event is signaled. By default, events have low occurrence values. However, you might find that a certain event indicates a problem only when it occurs repeatedly over an extended period of time. You can change the occurrence value assigned to that event so that the Availability Manager signals the event only when necessary. For example, suppose page fault spikes are common in your environment, and the Availability Manager frequently signals intermittent HITTLP, total page fault rate is high events. You could change the event's occurrence value to 3, so that the total page fault rate must exceed the threshold for three consecutive collection intervals before being signaled to the event log. To avoid displaying insignificant events, you can customize an event so that the Availability Manager signals it only when it occurs continuously.
Threshold	Most events are checked against only one threshold; however, some events have dual thresholds, where the event is triggered if either one is true. For example, for the LOVLSP, node disk volume free space is low event, the Availability Manager checks both of the following thresholds: Number of blocks remaining Percentage of total blocks remaining
Escalation actions	You can enter one or more of the following values: User: If the event occurs, the Availability Manager refers to the User Action field to determine what action to take. OPCOM: If the event occurs, and certain conditions are met (see Section 7.7), the Availability Manager passes that event to OPCOM. (Data Analyzer on OpenVMS only) HP OpenView: If the event occurs and certain conditions are met (see Section 7.7), the Availability Manager passes that event to HP OpenView. (OpenView agents must be installed and configured on the Data Analyzer node.)
User Action	When the Event escalation action field is set to User, User Action is no longer dimmed. You can enter the name of a procedure to be executed if the event displayed at the top of the page occurs. To use this field, see the instructions in Section 7.8.2.

The "Event explanation and investigation hints" section of the Event Customizations page, which is not customizable, includes a description of the event displayed and suggestions for how to correct any problems that the event signals.

7.8.2 Entering a User Action

Note

OpenVMS and Windows execute the User Action procedure somewhat differently, as explained in the following paragraphs.

The following notes pertain to writing and executing User Action commands or command procedures. These notes apply to User Actions on both OpenVMS and Windows systems.

The procedure that you specify as the User Action is executed in the following manner:
- It is issued to the operating system that is running the Availability Manager Data Analyzer.
- It is issued as a process separate from the one running the Availability Manager to avoid affecting its operation.
- It is run under the same account as the one running the Availability Manager Data Analyzer.
User Actions are intended to execute procedures that do not require interactive displays or user input.
You can enter User Actions for events on either a systemwide basis or a per-node basis:
- On a systemwide basis, the User Action is issued for an event that occurs on any node.
- On a per-node basis, the User Action is issued for an event that occurs only on a specific node.

If event logging is enabled, the Availability Manager writes events to the event log file (called AnalyzerEvents.log by default on OpenVMS systems and Windows systems). A status line matching the original line indicates whether the User Action was successfully issued. For example:

AMGR/KOINE -- 13-Apr-2005 15:33:02.531 --<0,CFGDON>KOINE configuration done AMGR/KOINE -- 13-Apr-2005 15:33:02.531 --<0,CFGDON>KOINE configuration done (User Action issued for this event on the client O/S)

Other events might appear between the first logging and the status line. The log file does not indicate whether the User Action executed successfully. You must obtain the execution status from the operating system, for example, the OpenVMS batch procedure log.

The User Action functionality might be enhanced in a future release of the Availability Manager, but backward compatibility is not guaranteed for the format of User Action procedure strings or for the method of executing the procedures on a particular operating system.

7.8.2.1 Executing a Procedure on an OpenVMS System

The User Action procedure and arguments are passed as string values to the DCL command interpreter as follows:

SUBMIT/NOPRINTER/LOG user_action_procedure arg_1 arg_2 arg_3 arg_4

where:

The first command is the DCL command SUBMIT with associated qualifiers.
user_action_procedure is a valid OpenVMS file name.
Enter the name of the procedure you want OpenVMS to execute using the following format:
disk:[directory]filename.COM
where:
- disk is the name of the disk where the procedure resides.
- directory is the name of the directory where the procedure resides.
- filename.COM is the file name of the command procedure you want OpenVMS to execute. The file name must follow OpenVMS file-naming conventions.
The file you submit must contain one or more DCL command statements that form a valid OpenVMS command procedure.

The arguments correspond to the following data:

Argument	Description
arg_1	Node name of the node that generated the event.
arg_2	Date and time that the event was generated.
arg_3	Name of the event.
arg_4	Description of the event.

The Availability Manager does not interpret the string contents. You can supply any content in the User Action procedure that DCL accepts in the OpenVMS environment for the user account running the Data Analyzer. However, if you include arguments in the User Action procedure, they might displace or overwrite arguments supplied by the Availability Manager.

A suitable batch queue must be available on the Data Analyzer computer to be the target of the SUBMIT command. See the HP OpenVMS DCL Dictionary for the SUBMIT, INITIALIZE/QUEUE, and START/QUEUE commands for use of batch queues and the queue manager.

The following is an example of a DCL command procedure:

$ if (p3.eqs."DSKOFF").and.(p1.eqs."PAYROL") $ then $ mail/subject="''p2' ''p3' ''p4'" urgent_instructions.txt call_center,finance,adams $ else $ mail/subject="''p2' ''p3' ''p4'" instructions.txt call_center $ endif

The pn numbers in the DCL procedure correspond in type, number, and position to the arguments in the preceding table.

You might use a procedure like this one to notify several groups if the payroll disk goes off line, or to notify the call center if any other event occurs.

7.8.2.2 Executing a Procedure on a Windows System

The user action procedure and arguments are passed as string values to the Windows command interpreter as follows:

"AT time CMD/C user_action_procedure arg_1 arg_2 arg_3 arg_4"

where:

AT is the Windows command that schedules commands and programs at a specified time and date.
The time substring is a short period of time--- aproximately 2 minutes---in the future so that the AT utility processes the user action procedure today rather than tomorrow. This is necessary because the AT utility cannot execute a procedure "now" rather than at an explicitly stated time.

user_action_procedure is a Windows command or valid file name. The file must contain one or more Windows command statements to form a valid command procedure. (See the example in this section.)
Enter the name of the procedure you want Windows to execute using the following format:
device:\(directory\filename.BAT
where:

device is the disk on which the procedure is located.
directory is the folder in which the procedure is located.

filename.BAT is the name of the command file to be executed.

Notes

The file name must follow Windows file-naming conventions. However, due to the processing of spaces in the Java JRE, HP recommends that you do not use spaces in a path or file name.
HP recommends that you use a batch file to process and call procedures and applications.

The arguments are listed in the table in Section 7.8.2.1.

The Availability Manager does not interpret the string contents. You can supply any content in the string that the Windows command-line interpreter accepts for the user account running the Data Analyzer. However, if you include arguments in the User Action procedure, they might displace or overwrite arguments supplied by the Availability Manager.

You cannot specify positional command-line switches or arguments to the AT command, although you can include switches in the User Action procedure substring as qualifiers to the user-supplied command. This is a limitation of both the Windows command-line interpreter and the way the entire string is passed from the Availability Manager to Windows.

The Schedule service must be running on the Data Analyzer computer in order to use the AT command. However, the Schedule service does not run by default. To start the Schedule service, see the Windows documentation for instructions in the use of the CONTROL PANEL->SERVICES->SCHEDULE->[startup button].

Windows Example

To set up a user action, follow these steps:

Select an event on the Event Customizations page, for example, HIBIOR (see Figure 7-20).
Change the Event escalation action to User.
Enter the name of the program to run. For example:
c:\send_message.bat

Figure 7-20 User Action Example

The command line parameters are automatically added when Availability Manager passes the command to the command processor.

The contents of "send_message.bat" are the following:

net send affc17 "P4:system event: %1 %2 %3 %4"

On the target node, AFFC17, a message similar to the following is displayed:

You can now apply the User Action to one node, all nodes, or a group of nodes, as explained in Section 7.8.2.

7.9 Customizing Security Features

The following sections explain how to change the following security features:

Passwords for groups and nodes
Data Analyzer passwords for OpenVMS and Windows Data Collector nodes
Security triplets on OpenVMS Data Collector nodes
Password on a Windows Data Collector node

Note

OpenVMS Data Collector nodes can have more than one password: each password is part of a security triplet. (Windows nodes allow you to have only one password per node.)

7.9.1 Customizing Passwords for Groups and Nodes

For both the Windows and OpenVMS Customization Pages at the operating system, group, or node level is a page similar to the one shown in Figure 7-6. It contains a tab labeled Security. If you select this tab on either system, the Availability Manager displays a page similar to the one shown in Figure 7-21.

Figure 7-21 OpenVMS Security Customization

The level at which you can make password changes depends on whether you select the Security tab at the operating system, group, or node level.

Changing Passwords at the Group Level

If you monitor several groups, but the password for the nodes in one of those groups is different from the password for nodes in other groups, right-click the group you want to change, select Customize from the list, select the Security tab, and change the password. The new password is then used for each node that is a member of that group.

Changing Passwords at the Node Level

As a second example, to change the password of one node in a group to a different password than the other nodes in the group, right-click that node, select Customize from the list, select the Security tab, and change the password to one that differs from the other nodes in the group. For that node, the new password overrides the group password.

In the second password example, if you want to set the password for the single node back to the password that the rest of the group uses, click Use default values. The password value for the node now comes from the group-level password setting. At this point, if you change the group password, all nodes in the group get the new password. Additional information about changing passwords for security is in Section 7.9.

7.9.2 Changing Data Analyzer Passwords

You can change the passwords that the Windows Data Analyzer uses for OpenVMS Data Collector nodes and for Windows Data Collector nodes. The following sections explain how to perform both actions.

7.9.2.1 Changing a Data Analyzer Password for an OpenVMS Data Collector Node

When you select Customize OpenVMS... on the Customize menu of the System Overview window, the Availability Manager displays a default customization page. On it is a tab marked Security, which, if you select it, displays the OpenVMS Security Customization page (Figure 7-21).

To change the default password for the Data Analyzer to use to access OpenVMS Data Collector nodes, enter a password of exactly 8 uppercase alphanumeric characters. The Availability Manager will use this password to access OpenVMS Data Collector nodes. This password must match the password that is part of the OpenVMS Data Collector security triplet ( Section 1.3.3).

When you are satisfied with your password, click OK. Exit the Availability Manager, and restart the application for the password to take effect.

7.9.2.2 Changing a Data Analyzer Password for a Windows Data Collector Node

When you select Customize Windows NT... on the Customize menu of the System Overview window, the Availability Manager displays a Windows Security Customization page (Figure 7-22).

Figure 7-22 Windows Security Customization

To change the default password for the Data Analyzer to use to access Windows Data Collector nodes, enter a password of exactly 8 alphanumeric characters. Note that this password is case sensitive; any time you type it, you must use the original capitalization.

This password must also match the password for the Windows Data Collector node that you want to access. (See Section 7.9.3 for instructions for changing that password.)

When you are satisfied with your password, click OK. Exit and restart the Availability Manager for the password to take effect.

7.9.3 Changing a Password on a Windows Data Collector

To change the Data Collector password in the Registry, follow these steps:

Click the Windows Start button. On the menu displayed, first select Programs, then Accessories, and then Command Prompt.
Type regedit after the angle prompt (>).
The system displays a screen for the Registry Editor, with a list of entries under My Computer.
On the list displayed, expand the HKEY_LOCAL_MACHINE entry.
Double-click SYSTEM.
Click CurrentControlSet.
Click Services.
Click damdrvr.
Click Parameters.
Double-click Read Password. Then type a new 8-character alphanumeric password, and click OK to make the change.
To store the new password, click Exit under File on the main menu bar.
On the Control Panel, select Services and then Stop for "PerfServ."
Again on the Control Panel, select Devices and then Stop for "damdrvr."
First restart damdrvr under "Devices" and then restart PerfServ under "Services."
This step completes the change of your Data Collector password.

7.10 Monitoring Processes on a Node

As the Availability Manager monitors all the processes on the system, you can configure the tool to notify you when particular processes are created or exit on your system. The Availability Manager can watch up to eight processes on an individual node or a node that is part of a group that you select on the Group/Node Inclusion Lists page (Figure 7-3). (You cannot, however, use this feature to notify you about processes that should not be there.)

To use the Watch Process feature, right-click a node in the Node pane to display the Node pane Customization Menu (Figure 7-23).

Figure 7-23 Node Customization Menu

On this menu, select Customize to display the options available for customizing individual nodes. One of these options is watch process, which you can select to display the Watch Process page (Figure 7-24).

Figure 7-24 Process Watch

An explanation of the watch process feature is displayed on the right side of the page. You can enter up to 8 processes in the box on the left side of the page. After you enter process names, the Availability Manager monitors these processes on the node that you have selected.

For a process that is not present on the node at the time you entered it on the Watch Process page, the Availability Manager displays the following event in the Event pane of the System Overview window (Figure 2-1):

NOPROC -- The process process-name has disappeared on the node node-name.

If a process that was signalled by a NOPROC event reappears on the node, the Availability Manager displays the following event in the Event pane of the System Overview window (Figure 2-1):

PRCFND -- The process process-name has recently reappeared on the node node-name.

Contents

Index