Table of Contents

Introduction to the Agent Manager 

Service Tab 

Database/Address Tab 

Database

Addresses

SNMP Tab

Notification Tab 

Advanced Tab 

SNMP Discovery

IP Server Discovery

General Discovery

About Tab 

Agent Events that Can be Logged to the Event Log

Close the Agent Manager's Dialog Box

Configure the Agent 

Configure the E-mail Feature 

Control What Problems Get Reported in the Console

Disable the Agent 

Make Changes Take Effect 

Overwrite or Append the Currently-Selected Database

See the Status of the Agent 

Select a Different Database to Store Information In

Set Global Threshold Levels 

Specify Alternate and RMON SNMP Community Strings

Specify the Default SNMP Community String

Specify the DNS Server's and Default Router's IP Addresses

Start the Agent Automatically When Windows Boots

Start the Agent 

Stop the Agent 

Use the Database with Standard Database Tools

Use the Event Viewer to View Agent Events

Problems Tab 

When Selecting the Problem Log Radio Button

When Selecting the Notifications Radio Button

What Devices the Agent Discovers

Ensure That the Agent Can Correctly Identify IPX Names

Service Dialog Box 

Discovery of Devices with Proprietary MIBs

The Agent's Device Discovery and Problem Reporting Processes

Passive Discovery Tests

Initial Active Discovery Tests

Ongoing Active Tests

Reporting of Errors, Warnings, and Information

Other Things that Affect When You See Information Change in the Console

Introduction to the Agent Manager

The Network InspectorT application consists of three main components:

The Agent Manager (also referred to as the Agent) discovers and analyzes information about the network.

A Microsoft Access ODBC relational database is where the Agent stores the information it has discovered about the network and its devices.

The Console is the Graphical User Interface (GUI) you can use to view the information collected by the Agent.

Refer to The Main Components of the Application in the Overview help for more details about the database and Console components.

The Windows NT/2000 Agent Manager runs as a service. The main function of the Agent is to discover and analyze information about the devices (servers, routers, switches, printers, managed hubs, hosts, and Fluke Networks tools) and the IP subnets, IPX networks, and NetBIOS domains on the local sub-network where the Agent resides. To do this, the Agent "listens" to the traffic on the sub-network, followed by sending broadcast and multicast packets and directed queries to devices on the sub-network.

When you start the Agent, it begins looking for devices on the network. The devices discovered by the Agent are limited to the devices that can be probed on the local side of the network, as bound by the first router(s) discovered. That is, the Agent can "see" all the devices on its local sub-network, up to and including a router, but it cannot "see" devices on the other side of a router. However, using the Console, you can add a device that is not on the Agent's local sub-network, and thereby cause the Agent to monitor a remote device as well. (Refer to Add a New Device in the Console help, for more information.)

Note

The exception to this is that all Fluke Networks Tools on your network will be discovered.

See Also

Refer to What Devices the Agent Discovers and The Agent's Device Discovery and Problem Reporting Processes in the Agent help, for more information on what devices the Agent discovers and how devices are discovered.

The Agent stores its discovery and analysis information in a Microsoft Access ODBC (Open DataBase Connectivity) relational database. The particular database in which the Agent's information is stored depends on the DSN specified in the Database Name field in the Database/Address Tab. By default, Inspector is specified in the Database Name field.

During installation, a default database, Netins.mdb, is created on the computer where the Agent is installed. This database is registered (with a Data Source Name of Inspector) with the Agent, so that the Agent knows where to find the default database. You can, however, select a different database for the Agent to store information in.

Note

You can also create a new database or create a copy of an existing database, register it and then select it to be the database for the Agent to store information.

The Windows NT/2000 Agent's user interface is a dialog box labeled Network Inspector Agent Manager, which contains five tabs:

The Service Tab lets you start and stop the Agent, and shows the current status of the Agent. It also provides the Startup. button, which opens the Service dialog box, where you can control when the Agent is started and whether the Agent is disabled. The Service dialog box also lets you specify an account for the Agent's service to run in, so that the Agent can properly determine IPX names when the Agent is monitoring an IPX network.

The Database/Address Tab lets you select a database, register a database, choose whether to overwrite the selected database, view the IP address and other IP configurations for the Agent, and select the network adapter to use (if there is more than one.).

Note

The Agent attempts to autodetect your machine's IP configuration information; however, certain situations limit the accuracy of this autodetection. Removal of TCP/IP, multiple NICs, moving from DHCP to fixed IP addressing (or vice-versa) may result in incorrect IP autodetection. If this happens, you must manually configure the IP for your machine in the Windows Control Panel. For more information, refer to your Microsoft documentation on using Network in Control Panel.

The SNMP Tab lets you set the value of various SNMP read-only Community Strings.

The Problems Tab lets you set the available global threshold levels and control which of the various types of errors (), warnings (), and information () that the Agent can detect will be saved in the database and reported in the Console (the Problem Log and Device Properties' Problems tab). When the Notifications radio button is selected, the Problems tab lets you control which of the various types of errors, warnings, and information that the Agent can detect that will also generate an e-mail and/or an e-mail to pager. For more information on notification, refer to the Notification Tab topic.

The Notification Tab lets you configure the Network Inspector application's Paging and/or E-mail notification features.

The Advanced Tab lets you enable or disable the Network Inspector application's SNMP discovery and IP Server discovery. By default, the SNMP discovery and IP Server discovery are enabled.

The About tab displays version and copyright information for the Agent Manager. It also shows serial number of the application.

The dialog box also has three buttons, which are available regardless of which tab you are in:

The Close button closes the Agent Manager's dialog box.

Note

Because the Windows NT/2000 Agent is a service, closing the Agent Manager's dialog box does not stop the Agent. If the Agent is running when you close the dialog box, it keeps running, gathering data, and storing data to the database until you click the Stop button in the Service Tab of the Agent Manager's dialog box.

The Apply button saves any changes you make in the Database/Address tab, the SNMP tab, the Problems tab, and the Advanced tab so that they take effect the next time you start the Agent.

The Help button causes this help file to be displayed.

Service Tab

The Service tab contains the following components:

The top of Status window shows the DSN that the agent is currently using. This is the same DSN that is selected on the Database/Address tab and is shown here as a quick reference. The Status field shows the current status of the Agent (usually Running or Stopped).

Notes

A status of Access Denied indicates that you do not have administration privileges on the computer where the Agent is installed. Service Not Installed indicates the Agent's computer was not rebooted after the Agent and the Windows NT/2000 driver were installed, so the Agent's service has not been activated.

Refer to Agent Events that Can be Logged to the Event Log for information on additional status messages that are reported as events in the Windows NT/2000 Event Log.

The Start button causes the Agent to start discovering and analyzing devices on the network.

The Agent puts its discovery and analysis information into the database that corresponds to the Data Source Name specified in the Database Name field in the Database/Address tab. The Agent either overwrites any information currently in the specified file, or the Agent uses any information already in the file as a baseline, depending on whether the Overwrite check box in the Database/Address tab is selected. A check in the Overwrite check box enables the overwrite mode.

Notes

The Agent can take up to 30 seconds to start.

If your computer does not have the required Microsoft TCP/IP stack installed you will get an error message stating that the stack is required in order to run the Network Inspector Agent.

The Agent does not collect any information about the network and its devices until you click the Start button. The Status message will indicate Running. Once the Agent is started, it continues to gather, analyze, and store network and device information until you click the Stop button. Closing the Agent Manager's dialog box (using the Close button or the in the upper right corner) does not stop the Agent.

You can start the Agent and leave the Agent running indefinitely, and just open and close the Console when you want to see the current status of your network and its devices.

The Stop button causes the Agent to stop discovering and analyzing devices on the network and to stop writing to the database. You can still use the Console to view the collected information.

The Startup. button opens the Service Tab, where you can control when the Agent is started and whether the Agent is disabled. It also lets you specify an account for the Agent's service to run in, so that the Agent can properly determine IPX names when the Agent is monitoring an IPX network.

To control when and how the Agent is started, click the Startup. button in the Service tab of the Agent's dialog box. The Service dialog box contains the following components:

The radio buttons in the Startup Type area let you control how and when the Agent is started:

If you select the Automatic radio button, the Agent will be started automatically the next time Windows NT/2000 is booted on the Agent's computer. If you select this option, make sure the parameters in the Database/Address Tab, the SNMP Tab, the Problems Tab, and the Advanced Tab are correct, because these parameters will be used when the Agent is started.

Notes

If you select the Automatic radio button, be aware that you are adding up to several minutes to the time it takes Windows NT/2000 to boot after you log in. You can still manually start the Agent whenever you want by clicking the Start button. The automatic-start option merely gives you a way to ensure that if the computer reboots, the Agent will be started again automatically.

If your computer does not have the required Microsoft TCP/IP stack installed you will get an error message stating that the stack is required in order to run the Network Inspector Agent.

If you select the Manual radio button, you can manually start the Agent by clicking the Start button. However, if the computer reboots, the Agent will not be automatically restarted.

If you select the Disabled radio button, the Agent cannot be started automatically or manually until you de-select the Disabled radio button.

The Log On As: area lets you specify an account for the Agent's service to run in. The Agent will use the specified account to gain access to the Novell server, so that the Agent can properly determine IPX names when the Agent is monitoring an IPX network.

The System Account radio button lets you select the Agent to run in the default System Account.

Note

If you select the Agent to run in the default System Account, you cannot monitor an IPX network. If you are not using the Agent to monitor an IPX network, leave the System Account radio button selected (enabled).

The This Account radio button and the This Account text field let you enter a user account.

The Password text field lets you enter a password.

Reenter the password in the Confirm Password field.

Note

The user account and password you enter must meet the following two criteria: They must be a valid NT account and password on the computer, and they must precisely match a valid Novell account and password.

Database/Address Tab

The Database/Address tab contains the following components:

Database

The Name field shows the DSN for the selected database that the Agent is using. Use the button to the right of the Name field to open the Click to select an already existing database that the Agent should use to store its discovery and analysis information.

The Manage. button opens the Manage Data Sources dialog box, where you can create a new database, assign a DSN to a database, copy a database and perform other database management tasks.

The Overwrite check box, if selected, causes the Agent, the next time it is started, to overwrite the database that corresponds to the DSN specified in the Name field. In other words, the Agent stores its new discovery and analysis information over any information already in the specified database, so the previous contents of the file are lost. You might want to overwrite the database, for example, if you made some major changes on your network and you want to "start over" by having the Agent re-discover the network and its devices.

Caution

If you enable overwrite, any notes in the database (which you entered in the Notes tab in the Console's Device Properties) will be lost. In addition, any new devices that you manually added to the database (using the Add New. item in the Console's Device menu) will no longer be in the database.

Note

Before you overwrite a database, make a copy of it using the Manage Data Sources dialog box. You can access it by clicking on the Manage. button.

If the Overwrite check box is not selected, the Agent uses the specified database as a baseline. That is, when the Agent discovers a device, it checks the database to see if the discovered device is already in the database. If it is, the Agent does not write information about that device to the database unless the device's information has changed. If the Agent discovers a device that is not already in the database, it writes the new device's information to the database.

If you are using the Network Inspector application as a long-term monitoring system (that is, you start the Agent and leave it running on a sub-network for a long period of time), you probably should not use the overwrite feature. Instead, you should have the Agent use the existing information in the database as a baseline.

Note

Regardless of being in the append mode, the interface statistics information of the optional Switch Inspector in the Console (and in some reports) will be cleared if the Agent has been stopped for 1 hour or longer.

Addresses

The IP Address field lets you view the IP address, in dotted-decimal format, for the Agent (for example, ). You cannot change this field. The Network Inspector application automatically manages the IP address to be the same as the IP address for the Agent's computer.

Notes

The Network Inspector application automatically manages addresses through the required Microsoft TCP/IP stack.

The Agent attempts to autodetect your machine's IP configuration information; however, certain situations limit the accuracy of this autodetection. Removal of TCP/IP, multiple NICs, moving from DHCP to fixed IP addressing (or vice-versa) may result in incorrect IP autodetection. If this happens, you must manually configure the IP for your machine in Windows' Control Panel. For more information, refer to your Microsoft documentation on using Network in Windows' Control Panel.

Also, if your are running a non-English version of Windows 98 you can enter your machine's IP configuration information in this screen.

The IP Subnet Mask field lets you view the IP subnet mask for the Agent. You cannot change this field. The Network Inspector application automatically manages the IP subnet mask to be the same as the IP subnet mask for the Agent's computer.

The DNS IP Address field lets you view the IP address of the DNS (Domain Name System) server the Agent will query when it attempts to discover a device's DNS name. When possible, the Network Inspector application automatically manages the DNS IP address to be the same as the DNS IP address for the Agent's computer. You cannot change this field from the Agent Manager's dialog box.

If the sub-network where the Agent resides does not have a DNS server, this field will be blank, but be aware that the Agent will not be able to resolve the IP addresses of devices to their DNS names.

Notes

Normally, if you have the required Microsoft TCP/IP stack properly running on your computer, you will not need to specify the DNS server's IP address. However, there are some situations where you would want to specify this address. In those cases that you need to change the DNS IP address for the Agent's computer, refer to Specify the DNS Server's and Default Router's IP Addresses for more information.

The Router IP Address field lets you view the IP address of a router that is on the sub-network where the Agent's computer resides. The Agent uses that router as its default router. When possible, the Network Inspector application automatically manages the router's IP address to be the same as the default router IP address for the Agent's computer. You cannot change this field from the Agent Manager's dialog box.

Notes

Normally, if you have the required Microsoft TCP/IP stack properly running on your computer, you will not need to specify the default router's IP address. However, there are some situations where you would want to specify this address. In those cases that you need to change the default router's IP address for the Agent's computer, refer to Specify the DNS Server's and Default Router's IP Addresses for more information.

The Network Adapter field lets you select the network adapter (if there are more than one) for the Agent to use.

SNMP Tab

The SNMP tab lets you enter SNMP community strings that the Agent uses when it queries a device to get SNMP and RMON information from the device's MIB. When the Agent discovers an IP device, it uses these strings in the following way:

The Agent first attempts to use the string specified in the Default field to get the device's SNMP System Group information. The default string in the Default field is public, which is the standard, read-only SNMP community string.

If the Agent was not able to query the device's SNMP agent using the Default string, the Agent does the following:

If the device is a key device, the Agent attempts to use the Alternate 1 string, the Alternate 2 string (if the Alternate 1 string fails), and then the Alternate 3 string (if the Alternate 2 string fails).

If the device is not a key device, the Agent stops trying to query the device.

If the Agent was able to query the device to get its SNMP System Group information, it does the following:

It queries the device again, using whatever string worked previously, to get the device's SNMP Interfaces Group information.

To find out if the device is an RMON device, the Agent queries the device again, using whatever string worked previously, to get any RMON-specific information in the device's MIB. Some devices use a unique community string for the RMON area of the MIB, so if using the previous string does not work, the Agent tries again using the string specified in the RMON field.

Notes

If the Agent is successful in getting RMON information, it uses this information rather than the Interfaces Group information to calculate interface utilization percentage and error percentage.

Any SNMP and RMON information the Agent succeeds in retrieving is displayed in the SNMP tab in the Console's Device Properties.

Notification Tab

The Notification tab lets you configure the E-mail and E-mail to Pager notification features. After configuring Notification, the Network Inspector application will notify you (by e-mail and/or e-mail to pager) each time there is a problem that is selected for notification. Problems are selected for notification in the Problems Tab.

The e-mail configuration contains the following components:

Note

In order for the E-mail feature to work, you must know your e-mail SMTP Server name.

The Enable E-mail Notification check box, when selected, enables e-mail configuration text fields and buttons.

The Outgoing SMTP Server: field lets you enter the DNS name or IP address of your SMTP e-mail server for the notification feature to use. A correct Outgoing SMTP Server is required.

"From" Account: field lets you enter an identifying email address (or string) so that the e-mail's source can be identified. The "From" Account field is optional. If you do not make an entry into this field, the computer name is used as the "From" Account text.

The Enter E-mail Address(es), then press "Add": field lets you enter an e-mail address for notification. After you add each e-mail address it is displayed in the Recipient column and you can indicate if it is a pager by selecting (putting a check in) the check box under the Pager column.

After entering text for an e-mail address, use the following buttons as needed:

The Add button lets you add the address to the E-mail Address list.

The Remove button lets you delete the selected address from the E-mail Address list.

The Remove All button lets you delete all addresses from the E-mail Address list.

The Max Message Length (for pagers only) field lets you enter the maximum character length for e-mail to pager messages, in decimal. If the e-mail destination is a pager or text messaging for a cell phone and your e-mail is too long for the device, your message will be truncated by the target device.

The Test E-mail button lets you send a test e-mail to all of the addresses on the E-mail Address list. A dialog window displays the results of the test operation.

The Set Notification Events button displays the Problems tab so that you can configure which problem events will activate an e-mail notification. Also refer to Control What Problems Generate a Notification.

See Also

Control What Problems Generate a Notification

Problems Tab

Advanced Tab

The Agent's discovery processes may occasionally expose a problem on some network devices with older software drivers and/or firmware (such as printers). If this happens, you should first verify that each problem device has the latest firmware and driver installed. If that does not fix the problem, refer to the information in this topic for ways to resolve the problem.

The Advanced tab lets you enable or disable the Network Inspector application's SNMP discovery, IP Server discovery (including DHCP servers), and large network discovery algorithms. By default, the SNMP discovery and IP Server discovery are enabled and the large network discovery algorithms is disabled.

Caution

If you disable IP Server discovery for an existing database, you may get erroneous IP Service No Longer Seen on Device changes.

Notes

SNMP discovery should almost always be enabled. Disable SNMP discovery only as a last resort to troubleshoot a network problem.

Disabling SNMP discovery prevents the optional Switch Inspector in the Console from displaying any new interfaces, MultiPort Statistics, or Port History. Disabling SNMP discovery also prevents the Network Inspector Agent from discovering SNMP Agents, utilization sources, interface information and statistics, some problems, and certain IP device types and/or services.

Disabling SNMP discovery and IP Server discovery will prevent information that would have normally been discovered by these means from being entered into the database and being displayed in the Console (including Network Inspector Reports).

To change any setting in the Advanced tab, you must first stop the Agent (if it is running), make the desired changes, "apply" the changes, and then re-start the Agent. To disable a particular type of discovery, make sure its check box is not selected (not checked). Conversely, to enable a particular type of discovery, make sure its check box is selected (checked). For more information on how to configure the Agent, refer to Configure the Agent.

SNMP Discovery

You must allow SNMP queries for the Switch Inspector option to function and to discover SNMP agents, Utilization Sources, interfaces, and some IP device types and services.

IP Server Discovery

In your effort to resolve an IP server discovery problem, you can disable all IP Server discovery or you can disable one or more of the following IP Server Discovery options:

Discover DHCP Servers

Discover HTTP Servers

Discover E-Mail Servers:

SMTP

POP2

POP3

IMAP4

General Discovery

If you have a large network or a smaller network with lots of broadcast traffic, you will need to apply the large network discovery algorithms in order for the Agent discovery to be optimized for this environment. (A large network is defined as more than 2000 network devices (including switches) or more than 75 switches).

When you select the Apply Large Network Discovery Algorithms, modifications are made to the way broadcast traffic is handled and the rediscovery wait time for your network is increased from 90 minutes to 12 hours.

See Also

Configure the Agent

Problems Tab

Switch Inspector Tab in the Console help

About Tab

The About tab displays version information, build, and serial numbers, valid options, and copyright information for the Network Inspector Agent. It provides a product registration link, other Fluke Networks product links and you can also click on the Fluke Networks logo to take you to the Fluke Networks home page.

Agent Events that Can be Logged to the Event Log

The following Network Inspector Agent events can be logged in the Windows NT/2000 system log and displayed in the Windows NT/2000 Event Viewer:

Note

For information on how to open the Event Viewer and how to view an event's detailed description, refer to Use the Event Viewer to View Agent Events.

Close the Agent Manager's Dialog Box

To close the Agent Manager's dialog box, click the in the upper right corner of the Agent Manager's dialog box, or click the Close button.

Note

Because the Windows NT/2000 Agent is a service, closing the Agent Manager's dialog box does not stop the Agent. If the Agent is running when you close the dialog box, it keeps running, gathering data, and storing data to the database. To stop the Agent and cause it to quit writing to the database, you must click the Stop button.

You can close the Agent Manager's dialog box, leave the Agent running indefinitely, and then just open and close the Console when you want to see the current status of your network and its devices.

Configure the Agent

To "configure" the Agent, you stop the Agent (if it is running), make the desired configuration changes, "apply" the changes, and then re-start the Agent.

1. If the Agent is running, stop the Agent by clicking the Stop button in the Service tab of the Agent's dialog box.

2. Use the Database/Address tab, the SNMP tab, the Problems tab, and the Advanced tab to configure various parameters, such as which database the Agent will store its information or to select which network adapter to use (if there is more than one.)

3. Use the Service Tab to control how and when the Agent will be started and also (if you are using the Agent to monitor an IPX network) to specify an account for the Agent's service to run in, so that the Agent can properly determine IPX names.

4. When you have made all the desired configuration changes, click the Apply button to save the changes.

The changes take effect the next time the Agent is started.

The Network Inspector application automatically manages the Agent's IP address, IP subnet mask, DNS IP address, and Router IP address to be the same as the Agent's computer. You cannot change these fields from the Agent dialog box. The Network Inspector application automatically manages these addresses through the required Microsoft TCP/IP stack.

Notes

The Agent attempts to autodetect your machine's IP configuration information; however, certain situations limit the accuracy of this autodetection. Removal of TCP/IP, multiple network adapters, moving from DHCP to fixed IP addressing (or vice-versa) may result in incorrect IP autodetection. If this happens, you must manually configure the IP for your machine in the Windows Control Panel. For more information, refer to your Microsoft documentation on using Network in the Windows Control Panel.

If you are running a non-English version of Windows 98/Me, you must manually enter your machine's IP configuration information in this screen.

If you need to change the DNS server's or default router's IP address for the Agent's computer, refer to Specify the DNS Server's and Default Router's IP Addresses.

Configure the E-mail Feature

To configure the E-mail feature you need to configure Notification for E-mail and to specify which problem events should generate an e-mail notification and which events should generate an e-mail to pager notification. Use the following procedure to configure Notification for E-mail or Notification for E-mail to Pager:

Notes

In order for the E-mail feature to work, you must know your e-mail SMTP Server name.

If the e-mail destination is a pager or text messaging for a cell phone and your e-mail is too long for the device, your message will be truncated by the target device.

1. If the Agent is running, stop the Agent by clicking the Stop button in the Agent tab of the Agent Manager's dialog box.

2. In the Notification Tab, select (check) the Enable E-mail Notification check box.

3. Enter the DNS name or IP address of your SMTP e-mail server into the Outgoing SMTP Server field.

4. Enter an email address (or string) in the "From" Account to identify the e-mail's source (optional). If you do not make an entry into this field, the computer name that the Agent is running on is used as the "From" Account text.

5. Enter an e-mail address for notification into the Enter E-mail Address(es), then press "Add": field and select the Add button. After you add each e-mail address, it is displayed in the Recipient column.

a) If you need to delete an address from the E-mail Address list, highlight the address and then press dthe Remove button.

b) If you need to delete all addresses from the E-mail Address list use the Remove All button.

6. For each e-mail address that is a pager, select (check) the check box under the Pager column.

7. Enter the maximum character length for the e-mail to pager message, in decimal, into the Max Message Length field. If the e-mail destination is a pager or text messaging for a cell phone and your e-mail is too long for the device, your message will be truncated by the target device.

8. Select the Set Notification Events button and configure which problem events you want to activate an e-mail notification and/or an e-mail to pager notification. For more information, refer to Control What Problems Generate a Notification.

9. Send a test e-mail to all of the addresses on the E-mail Address list by selecting the Test E-mail button. A dialog window displays the results of the test operation.

10. Click the Apply button to save your changes.

The changes take effect the next time the Agent is started.

See Also

Notification tab

Problems Tab

Control What Problems Generate a Notification

Control What Problems Get Reported in the Console

When you select the Problem Log radio button in the Problems tab you can select which errors (), warnings (), and information () that the Agent can detect will be saved in the database and reported in the Console (the Problem Log and Device Properties Problems tab).

1. If the Agent is running, stop the Agent by clicking the Stop button in the Agent tab of the Agent Manager's dialog box.

2. In the Problems tab, verify that the Problem Log radio button is selected, click the Errors category, and select (check) the errors that you want reported.

Note

The error, Incorrect Configuration of Agent is always reported to the Problem Log, so there is no check box for it.

3. Repeat step 2 using the Warnings category and then the Information category to enable or disable the reporting of the warnings and information.

4. Click the Apply button to save any changes.

The changes take effect the next time the Agent is started.

See Also

Problems Tab

Control What Problems Generate a Notification

Disable the Agent

The three radio buttons in the Startup Type area of the Service dialog box let you control how and when the Agent is started. If you select the Disabled radio button, the Agent cannot be started automatically or manually until you de-select the Disabled radio button.

Make Changes Take Effect

Whenever you change any of the parameters in the Database/Address tab, the SNMP tab, or the Problems tab, the Apply button is enabled (that is, it is no longer grayed-out). You must click the Apply button to save the configuration changes. The saved changes then take effect the next time you start the Agent.

Overwrite or Append the Currently-Selected Database

When you start the Agent, the Agent puts its discovery and analysis information into the database that corresponds to the Data Source Name specified in the Database Name field in the Database/Address tab.

If the Overwrite check box in the Database/Address tab is selected, the Agent's new discovery and analysis information is written over any information already in the specified database, so the previous contents of the database are lost.

Note

Before you overwrite a database, you can copy it.

Caution

If you enable overwrite, any notes in the database (which you entered in the Notes tab of the Console's Device Properties) will be lost. In addition, any new devices that you manually added to the database (using the Add New. item in the Console's Device menu) will no longer be in the database.

If the Overwrite check box is not selected, the Agent will use the information already in the specified database as a baseline and append any new information. That is, when the Agent discovers a device, it checks the database to see if the discovered device is already in the database. If it is, the Agent does not write information about that device to the database unless the device's information has changed. If the Agent discovers a device that is not already in the database, it writes the new device's information to the database.

Note

Regardless of being in the append mode, the interface statistics information of the optional Switch Inspector in the Console (and in some reports) will be cleared if the Agent has been stopped for 1 hour or longer.

See Also

Database/Address Tab

See the Status of the Agent

To see the current status of the Agent (usually Running or Stopped), look at the Status field in the Service tab of the Agent Manager's dialog box.

Notes

Access Denied indicates that you do not have administration privileges on the computer where the Agent is installed.

Service Not Installed indicates the Agent's computer was not rebooted after the Agent and the Windows NT/2000 driver were installed, so the Agent's service has not been activated.

Refer to Agent Events that Can be Logged to the Event Log for information on additional status messages that are reported as events in the Windows NT/2000 Event Log.

Select a Different Database to Store Information In

You can select a different, already-existing database for the Agent to store its discovery and analysis information in from the Database/Address tab.

Note

You can also create a new database using the Manage Data Sources dialog box and then use the procedure below to select it.

If the Agent is running, stop the Agent by clicking the Stop button in the Service tab of the Agent Manager's dialog box.

In the Database/Address tab, click the button.

The Select Data Source dialog box opens.

In the Data Source Name window, select the Data Source Name that corresponds to the database you want the Agent to use.

Click Open.

The Open Data Source dialog box closes and the selected Data Source name is put into the Database Name field in the Database/Address tab.

Select the Overwrite check box if you want the Agent to overwrite any information already in the specified database.

OR

If you want the Agent to use any information in the specified database as a baseline, make sure the Overwrite check box is not selected (not checked). This causes the Agent, when it discovers a device, to check the database to see if the discovered device is already in the database. If it is, the Agent does not write information about that device to the database unless the device's information has changed. If the Agent discovers a device that is not already in the database, it writes the new device's information to the database.

Note

Before you overwrite a database, you can make a copy of it using the Manage Data Sources dialog box.

Caution

If you enable overwrite, any notes in the database (which you entered in the Notes tab in the Console's Device Properties) will be lost. In addition, any new devices that you manually added to the database (using the Add New. item in the Console's Device menu) will no longer be in the database.

Click the Apply button to save these changes.

The changes take effect the next time the Agent is started.

Set Global Threshold Levels

You can set the global threshold levels for Utilization, Errors, and Collisions from the Default Interface Thresholds dialog box. Do the following to set global threshold levels:

1. If the Agent is running, stop the Agent by clicking the Stop button in the Agent tab of the Agent's dialog box.

2. In the Problems tab, click the Thresholds. button.

This opens the Default Interface Thresholds dialog box.

3. Click the Custom radio button of the threshold type that you want to set.

This enables the Error % and Warning % text fields.

4. Enter your threshold value (from 1% to 100%) into the Error % or Warning % text field.

5. Repeat steps 3 and 4 to set additional threshold setting.

6. Click the OK button to accept these changes.

7. Click the Apply button, in the Problems tab, to save these changes.

The new threshold levels take effect the next time you start the Agent.

The Agent logs to the database any errors or warnings generated due to these new thresholds for reporting in the Console (the Problem Log and Device Properties Problems tab).

See Also

Problems Tab

Set Individual-Interface Threshold Levels in the Console help

Specify Alternate and RMON SNMP Community Strings

You can use the three Alternate fields and the RMON field in the SNMP tab to specify alternate read-only SNMP Community Strings for the Agent to use if it is unable to query a device's SNMP agent using the string specified in the Default field.

Note

For information on how and when the Agent uses these community strings refer to SNMP Tab.

1. If the Agent is running, stop the Agent by clicking the Stop button in the Service tab of the Agent Manager's dialog box.

2. In the SNMP tab, enter the desired text into the Alternate fields and the RMON field.

3. Click the Apply button to save these changes.

The changes take effect the next time the Agent is started.

See Also

Specify the Default SNMP Community String

Specify the Default SNMP Community String

You can use the Default field in the SNMP tab to specify the first string the Agent should use when it attempts to query a device's SNMP agent to get its SNMP System Group and Interfaces Group information and RMON information.

1. If the Agent is running, stop the Agent by clicking the Stop button in the Service tab of the Agent Manager's dialog box.

2. In the SNMP tab, enter the desired text into the Default field.

3. Click the Apply button to save this change.

The change takes effect the next time the Agent is started.

See Also

SNMP Tab

Specify Alternate and RMON SNMP Community Strings

Specify the DNS Server's and Default Router's IP Addresses

Use Network in Windows' Control Panel to specify the DNS server's IP address or the default router's IP address for your computer. Your computer's default router is also known as a gateway.

Refer to your Microsoft documentation for information on using Windows' Control Panel.

Notes

Normally, if you have the required Microsoft TCP/IP stack properly running on your computer, you will not need to specify the DNS server's IP address or the default router's IP address. However, there are some situations where you would want to specify these addresses.

Refer to your Microsoft documentation for information on running the Microsoft TCP/IP stack.

Start the Agent Automatically When Windows Boots

Using the Startup Type area in the Service dialog box, you can set up the Agent so that it is started whenever Windows boots on the Agent's computer.

Notes

If you set up the Agent so that it starts automatically when Windows boots, be aware that you are adding up to 30 seconds to the time it takes the PC to boot after you log in. Also, you can still manually start the Agent whenever you want by clicking the Start button. The automatic-start option merely gives you a way to ensure that if the computer reboots, the Agent will be started again automatically.

If your computer does not have the required Microsoft TCP/IP stack installed you will get an error message stating that the stack is required in order to run the Agent.

You can sometimes get the following error if you start the Agent just before opening the Console. The error occurs because the Agent compacts its database when it is started and this prevents the Console from opening. You can clear this condition by closing the error popup, waiting about 30 seconds, and then restarting the Console. If you always start the Agent from the console, you should never see this error.

ODBC Microsoft Access 2000 Driver Login Failed

Couldn't use "|"; file already in use

If you get the ODBC Tracing is Enabled warning while starting the Agent or the Console; you should disable ODBC Tracing to obtain maximum performance from your computer. The Network Inspector application disables ODBC Tracing (even if it is enabled globally) for all of its processes except reports. After you close the warning and allow the Agent or the Console to open, you can disable ODBC Tracing from Window's Control Panel, ODBC Data Sources, the Tracing tab, and select the Stop Tracing Now button. To continue without disabling ODBC Tracing, simply close the warning and then the Agent or the Console will continue to open and operate.

1. In the Service tab of the Agent Manager's dialog box, click the Startup. button.

The Service dialog box opens.

2. Select the Automatic radio button.

3. Click OK.

Note

Make sure the rest of the parameters in the Database/Address tab, the SNMP tab, the Problems tab, and the Advanced tab are correct, because these parameters will be used if the Agent is automatically started.

See Also

Configure the Agent

Start the Agent

Starting the Agent causes it to start discovering and analyzing devices on the network and to store its discovery and analysis information in the database.

Cautions

You should never start the Agent until you check which database is specified in the Database/Address tab. If you start the Agent without checking the Database/Address tab you might inadvertently overwrite the database.

The packet drivers that the Agent uses can conflict with other products that use the packet driver at the same time. This can cause operational problems or incomplete data collection. You may have to close other applications if this problem occurs.

Notes

You can sometimes get the following error if you start the Agent just before opening the Console.

ODBC Microsoft Access 2000 Driver Login Failed

Couldn't use "|"; file already in use

The error occurs because the Agent compacts its database when it is started and this prevents the Console from opening. You can clear this condition by closing the error popup, waiting about 30 seconds, and then restarting the Console. If you always start the Agent from the console, you should never see this error.

If you get the ODBC Tracing is Enabled warning while starting the Agent or the Console; you should disable ODBC Tracing to obtain maximum performance from your computer. The Network Inspector application disables ODBC Tracing (even if it is enabled globally) for all of its processes except reports. After you close the warning and allow the Agent or the Console to open, you can disable ODBC Tracing from Window's Control Panel, ODBC Data Sources, the Tracing tab, and select the Stop Tracing Now button. To continue without disabling ODBC Tracing, simply close the warning and then the Agent or the Console will continue to open and operate.

In the Database/Address tab of the Agent Manager's dialog box, make sure the Database Name field contains the DSN that corresponds to the database you want the Agent to write to. If it does not, use the the button to the right of the Name field to open the Manage Data Sources dialog box to select an already existing database that the Agent should use to store its discovery and analysis information.

If you want to create a new database, use the Manage. button to open the Manage Data Sources dialog box, where you can create a new database, assign a DSN to a database, copy a database and other database management tasks.

2. Also in the Database/Address tab, decide whether you want the Agent to overwrite or append the database, and set the Overwrite check box accordingly.

Note

Before you overwrite a database, you can use the Manage Data Sources dialog box to make a copy.

Caution

If you enable overwrite, any notes in the database (which you entered in the Notes tab in the Console's Device Properties) will be lost. In addition, any new devices that you manually added to the database (using the Add New. item in the Console's Device menu) will no longer be in the database.

3. In the SNMP tab, specify the Default SNMP Community String and, optionally, specify the Alternate and RMON SNMP Community Strings.

4. In the Problems tab, use the check boxes to specify which errors (), warnings (), and information () should be saved in the database and reported in the Console (the Problem Log and the Device Properties' Problems tab)Control_What_Problems_Get_Reported_in_the_Console>(w95sec).

5. If you made any changes in the Database/Address tab, the SNMP tab, or the Problems tab, click the Apply button to save the changes.

6. In the Service tab, click the Startup. button.

The Service dialog box opens.

7. In the Service dialog box, do the following:

a) Select the Automatic radio button if you want to ensure that the Agent will be started automatically the next time Windows NT/2000 is booted on the Agent's computer, or select the Manual radio button if you do not want the Agent to be started automatically the next time Windows NT/2000 is booted.

Notes

If you select the Automatic radio button, be aware that you are adding up to 30 seconds to the time it takes Windows NT/2000 to boot after you log in.

Regardless of whether you select the Automatic radio button or the Manual radio button, you can still manually start the Agent whenever you want by clicking the Start button. The automatic-start option merely gives you a way to ensure that if the computer reboots, the Agent will be started again automatically.

b) If you are using the Agent to monitor an IPX network, use step c through step f below to specify an account for the Agent's service to run in, so that the Agent can properly determine IPX names; otherwise, go to step f.

Note

If you are not using the Agent to monitor an IPX network, leave the System Account radio button selected (enabled).

c) Click the This Account radio button and then in the This Account text field, enter a user account.

d) In the Password text field, enter a password.

e) In the Confirm Password text field, enter the same password you entered in the Password text field.

Note

The user account and password you enter must meet the following two criteria: they must be a valid NT account and password on the computer, and they must precisely match a valid Novell account and password.

f) Click OK.

The Service dialog box closes.

8. In the Service tab, click the Start button to start the Agent.

The Status field displays Running.

Notes

If your computer does not have the required Microsoft TCP/IP stack installed you will get an error message stating that the stack is required in order to run the Network Inspector Agent.

The Status field displays Service Not Installed if you did not activate the service by rebooting after you installed the application's driver. It displays if Access Denied you do not have administration privileges.

Once the Agent is started, it continues to gather network and device information until you click the Stop button. Closing the Agent Manager's dialog box (using the Close button or the in the upper right corner) does not stop the Agent.

You can start the Agent and leave the Agent running indefinitely, and just open and close the Console when you want to see the current status of your network and its devices.

Stop the Agent

To stop the Agent, click the Stop button in the Service tab of the Agent Manager's dialog box. When you click the Stop button, the database is saved and the Status field in the Service tab displays Stopped.

Notes

Because the Windows NT/2000 Agent is a service, using the Close button or the in the upper right corner does not stop the Agent (although these buttons do close the Agent Manager's dialog box). Closing the Console does not stop the Agent either.

Stopping the Agent will interrupt any trend data that you might be collecting, and stopping the Agent for more than 1 hour will clear all trend data.

Use the Database with Standard Database Tools

The Agent stores its discovery and analysis information in a Microsoft Access ODBC (Open DataBase Connectivity) relational database. You can use standard database tools like Microsoft Access 2000 or Seagate Crystal Reports version 8.5 with the Network Inspector database. To do this, you will need proficiency with the following:

Microsoft Access ODBC (Open DataBase Connectivity) relational database

Database schemas and how they relate to the Network Inspector database

The database tool that you select to use

For example, once you open a Network Inspector database in Microsoft Access 2000, you can then do all the database-related tasks that you can normally do in Access, such as create queries and forms and generate reports. You could, for example, create a query to show you all the problems that occurred after a certain time, or a query to show you all the devices that have interfaces that are down. For more information, refer to your Microsoft Access 2000 documentation.

Cautions

If you open a Network Inspector database in another database tool, be very careful not to change any of the data in the database. If any of the data is changed, you may run into problems using the database with the Network Inspector application. Therefore, it is a good idea to make a copy of the database before you use it with another database tool.

If you use the default database file (Netins.mdb) without copying it and it is then corrupted when you use it in another tool, you can use Windows Explorer to copy the file Empty.mdb (found in Program Files\Fluke Networks\Network Inspector\Database if you used the default directory during installation) to Netins.mdb. Your other option is to reinstall the application.

The Network Inspector database files are compatible with Microsoft Access 2000. If you attempt to open a database file with a newer version of Microsoft Access, you will get a message indicating the file needs to be updated. However, if you update the file, it will no longer work with the Network Inspector application.

The data schema for the Network Inspector application is not guaranteed to remain the same in future revisions.

Notes

If you use a standard database tool to compact the Network Inspector database, the database file must not be accessed by any application at the time.

If you selected the default installation directory during installation of the application, the Network Inspector database files will be in the C:\Program Files\Fluke Networks\Inspector\Database directory.

See Also

Network Inspector Reports that You can Run

LANMeter Instrument Reports

Use the Event Viewer to View Agent Events

You can use the Windows NT/2000 Event Viewer to view Agent events that are recorded in the Windows NT/2000 system log. For example, when you start or stop the Agent, an event is recorded in the system log. In addition, if the Agent's service is unable to start, an event is recorded that describes why the service could not start.

1. To open the Event Viewer, click the Windows Start button.

2. Point to Programs.

3. Point to the Administrative Tools.

4. Click Event Viewer.

The Event Viewer window opens, displaying events in chronological order, with the most recent event displayed at the top of the Event Viewer.

5. If the system log is not displayed, select the System item in the Log menu.

6. Press F5, which refreshes the display, to ensure all events are displayed.

7. Double-click an event to open the Event Detail dialog box, where you can see a description of the event.

Refer to Agent Events that Can be Logged to the Event Log for a list and description of the Agent events that can be displayed in the Event Viewer.

Problems Tab

The Problems tab has two modes, which are selectable by the Problem Log radio button and the Notifications radio button.

When Selecting the Problem Log Radio Button

When Selecting the Notifications Radio Button

When Selecting the Problem Log Radio Button

The Problems tab lets you set the available global threshold levels and control which types of errors (), warnings (), and information () that the Agent can detect will be saved in the database and reported in two places in the Console:

In the Problem Log

In the Problems tab of the Console's Device Properties dialog box

By default, all of the check boxes are selected except the Duplicate NetBIOS Name error, the two IP Service changes, and the Overlapped Subnet Mask error. All occurrences of the selected items will be saved in the database and reported in the Console. To enable the reporting of an error, warning, or change, make sure its check box is selected (checked). Conversely, to disable the reporting of an error, warning, or change, make sure its check box is not selected (not checked).

The Duplicate NetBIOS Name error is disabled by default due to the possibility of generating many Duplicate NetBIOS Name errors on networks with a lot of docking stations for laptop computers.

The two IP Service changes are disabled due to the possibility of generating many IP Service has Resumed on Device and IP Service No Longer Seen on Device changes due to response delays from devices that the application is monitoring. The application uses passive means to determine the on-going status of IP services after initially discovering them. Since some services broadcast their status but others don't, situations can arise where the Agent thinks a service is not available when the service is still available and vice versa. For example, if, during the interval that the Agent is monitoring the network for a service's status, the Agent does not see any packets indicating the service is still available, the Agent will think the service is no longer available.

Caution

If you enable the reporting of these changes, it is important that you understand that if the IP Service No Longer Seen on Device change is ever reported to the Problem Log, you should interpret this to mean that the service MAY have gone down, and you may want to investigate further. You should NOT interpret this change to mean that the service definitely went down. You can use the Re-discover All command in the Device menu to verify if it truly went down.

Note

The error, Incorrect Configuration of Agent is always reported to the Problem Log, so there is no check box for it.

Clicking a expands a category so that the categories beneath it are shown. Similarly, clicking a collapses the category so that the subcategories beneath it are not shown. Also, a white check box with a check in it means that all of the items in that category are selected (checked) and a gray check box with a check in it means that the some, but not all, of the items in that category are selected (checked).

The Problems tab also provides access to the Default Interface Thresholds dialog box, which you can use to set the global threshold levels for Utilization, Errors, and Collisions. You can set a global threshold by pressing the Thresholds. button to open the Default Interface Thresholds dialog box, clicking on the Custom radio button of the threshold type that you want to set, and then entering your threshold value into the Error % or Warning % text field. The new threshold levels take effect for the next set of statistics monitored. The Agent logs to the database any errors or warnings generated due to these new thresholds for reporting in the Console.

The following is the Default Interface Thresholds dialog box:

The following are the global threshold parameters and their default values (in parentheses):

Utilization

Error % (80)

Warning % (50)

Errors

Error % (1)

Collisions

Error % (20)

Warning % (5)

See Also

Set Individual-Interface Threshold Levels in the Console help

Control What Problems Get Reported in the Console

When Selecting the Notifications Radio Button

The Problems tab lets you control which of the various types of errors (), warnings (), and information () that the Agent can detect that will also generate an e-mail and/or an e-mail to pager.

After selecting the Notifications radio button, select either Standard E-mail or E-mail to Pager in the text field as the method of notification. Then you can select the various types of errors (), warnings (), and information () that you want to generate that type of notification. To select an error, warning, or change for an e-mail (or an e-mail to pager) notification, make sure its check box is selected (checked). Conversely, to not select an error, warning, or change, make sure its check box is not selected (not checked). If you want to activate both e-mail and e-mail to Pager notifications, you also need to select the second method of notification (Standard E-mail or E-mail to Pager) and then select the errors, warnings, and information that it should respond to.

By default, only the Key device not responding error check box is selected for both Standard E-mail and E-mail to Pager. You can make different selections for e-mail and e-mail to pager notifications. For example, you can be e-mailed to your pager when any one of only two errors occur and e-mailled when any one of four errors or two warnings occur.

See Also

Notification tab

Control What Problems Generate a Notification

What Devices the Agent Discovers

The Network Inspector application discovers servers, routers, switches, printers, managed hubs, clients (hosts), and Fluke Networks Tools.

Note

For a description of what types of servers, routers, switches and printers can be discovered and displayed in the Console's Explorer View, refer to Definition of Device Categories in the Left Pane in the Console help.

The following rules determine what devices the Agent will discover or "see":

The Agent discovers the devices in the local broadcast domain where the computer that is running the Agent resides. "Sub-network" in the context of the Network Inspector application is defined as the set of all devices that belong in a MAC.

The application has two operational modes: local and remote. The installation process is different for each operational mode, and the operational mode has the following implications on what devices the Agent discovers:

If you do a local operation installation of the application (where the Agent and the Console are installed on the same computer), the Agent discovers the devices in the sub-network where that computer resides.

If you do a remote operation installation (where one or more Agents are installed on separate computers than the one running the "local" Console/Agent.), the remote Agent discovers the devices on the sub-network where the remote Agent's computer resides. However, if the computer on which the Console/Agent is installed is in a different sub-network, the remote Agent will not discover the devices in the sub-network where the Console's computer resides.

The devices discovered by the Agent are limited to the devices that can be probed on the local side of the broadcast domain, as bound by the first router(s) discovered. That is, the Agent can see all the devices in the local sub-network, up to and including a router, but it cannot see devices on the other side of a router (although, using the Console, you can add a device that is not on the Agent's local sub-network, and thereby cause the Agent to monitor a remote device as well).

One exception to only discovering devices in the local broadcast domain is that all Fluke Networks tools on your network will be discovered. Each discovered Fluke Networks tool is first discovered as an IP host in the same manner as other IP devices. When a device has been discovered and entered into the database, a check is made to determine if it has a Fluke MAC address. If it does, it is tagged as a Fluke Networks tool. Furthermore, when the application discovers an SNMP managed router, its ARP table is examined. Any devices in the database that have Fluke MAC addresses are also tagged as Fluke Networks tools.

Note

You can deploy multiple remote software agents or hardware agentson your network and successively load their databases in the Console to get a complete view of your network.

The Agent might not discover some of the devices in the local broadcast domain:

Passive devices, such as bridges and hubs, are not discovered by the Agent, although non-passive bridges can be discovered.

Some switches do not "talk" much, and, therefore, might not be discovered. However, just as you can add a remote device to the Agent's database, you can add such a switch to the Agent's database, and thereby cause the Agent to monitor it as well.

When there are multiple IP subnets in the broadcast domain, some devices that are on subnets other than the one that the Agent is on will not be discovered. You can add these devices to the Agent's database, and thereby cause the Agent to monitor it as well.

If the Agent discovers UNIX servers, RMON probes, SQL servers, or non-passive bridges, these devices will be displayed under the Hosts category in the "tree" in the left pane of Console's Explorer View. (Any of these that run an SNMP agent and replied to at least the Agent's query for SNMP System Group information will also be listed under the SNMP Agents category.)

The rules above are illustrated in the figure below, which shows a network in which there are three sub-networks or broadcast domains:

If you did a local-operation install of the application on the Administrator's PC, the Agent would "see" the devices in Broadcast Domain B.

If instead, you did a remote-operation install in which you installed the Agent on the Production Server and the Console on the Administrator's PC, because the Production Server and the Administrator's PC are in different sub-networks, the Agent would see only the devices in Broadcast Domain C.

See Also

The Agent's Device Discovery and Problem Reporting Processes

Discovery of Devices with Proprietary MIBs

Ensure That the Agent Can Correctly Identify IPX Names

If you are using the Agent to monitor an IPX network, use the procedure below to specify an account for the Agent's service to run in. The Agent will use the specified account to gain access to the Novell server, so that the Agent can properly determine IPX names.

1. In the Service tab of the Agent Manager's dialog box, click the Startup. button.

The Service dialog box opens.

2. Click the This Account radio button and then in the This Account text field, enter a user account.

3. In the Password text field, enter a password.

4. In the Confirm Password text field, enter the same password you entered in the Password text field.

Note

The user account and password you enter must meet the following two criteria: They must be a valid NT account and password on the computer, and they must precisely match a valid Novell account and password.

5. Click OK.

The Service dialog box closes.

Control What Problems Generate a Notification

When the Notifications radio button is selected, the Problems tab lets you control which of the various types of errors (), warnings (), and information () that the Agent can detect that will also generate an e-mail and/or an e-mail to pager.

Notes

By default, only the Key device not responding error check box is selected for both e-mail and e-mail to pager. You can make different selections for e-mail and e-mail to pager notifications. For example, you can send e-mail to a pager when any one of only two errors occur and you can send standard e-mail when any one of four errors or two warnings occur.

In order for the E-mail feature to work, you must know your e-mail SMTP Server name.

1. If the Agent is running, stop the Agent by clicking the Stop button in the Agent tab of the Agent Manager's dialog box.

2. In the Problems tab, verify that the Notifications radio button is selected and then select Standard E-mail or E-mail to Pager in the text field. (If you want to control what problems get an e-mail notification select Standard E-mail, or if you want to control what problems get an e-mail to pager notification select E-mail to Pager).

3. Click the Errors category, make sure the check boxes that correspond to the errors that you want to generate a notification contain a check mark and make sure the check boxes that correspond to the errors that you do not want to generate a notification do not contain a check mark.

4. Repeat step 3 using the Warnings category and then the Changes category to enable or disable the notification response for warnings and information.

5. Repeat this procedure for the second mode of notification (optional). (For example, if you have just selected all of the problem check boxes that you want to generate an e-mail to pager notification, you can repeat this procedure to select all of the problem check boxes that you want to generate an e-mail notification).

6. Click the Apply button to save any changes.

The changes take effect the next time the Agent is started.

See Also

Notification tab

Problems Tab

Control What Problems Get Reported in the Console

Service Dialog Box

To open the Service dialog box, click the Startup. button in the Service tab of the Agent Manager's dialog box. The Service dialog box contains the following components:

The radio buttons in the Startup Type area let you control how and when the Agent is started:

If you select the Automatic radio button, the Agent will be started automatically the next time Windows is booted on the Agent's computer. If you select this option, make sure the parameters in the Database/Address tab, the SNMP tab, the Problems tab, and the Advanced tab are correct, because these parameters will be used when the Agent is automatically started.

Notes

If you select the Automatic radio button, be aware that you are adding up to several minutes to the time it takes Windows 2000 or Windows NT to boot after you login. Also, you can still manually start the Agent whenever you want by clicking the Start button. The automatic-start option merely gives you a way to ensure that if the computer reboots, the Agent will be started again automatically.

If your computer does not have the required Microsoft TCP/IP stack installed you will get an error message stating that the stack is required in order to run the Network Inspector Agent.

If you select the Manual radio button, you can manually start the Agent whenever you want by clicking the Start button. However, if the computer reboots, the Agent will not be automatically restarted.

If you select the Disabled radio button, the Agent cannot be started automatically or manually until you de-select the Disabled radio button.

The Log On As: area lets you specify an account for the Agent's service to run in. The Agent will use the specified account to gain access to the Novell server, so that the Agent can properly determine IPX names when the Agent is monitoring an IPX network.

The System Account radio button lets you select the Agent to run in the default System Account.

Note

If you select the Agent to run in the default System Account, you cannot monitor an IPX network. If you are not using the Agent to monitor an IPX network, leave the System Account radio button selected (enabled).

The This Account radio button and the This Account text field let you enter a user account.

The Password text field lets you enter a password.

The Confirm Password text field lets you enter the same password.

Note

The user account and password you enter must meet the following two criteria: They must be a valid NT account and password on the computer, and they must precisely match a valid Novell account and password.

Discovery of Devices with Proprietary MIBs

The Agent uses SNMP queries to obtain MIB-2 information from switches that support MIB-2. Some devices do not support MIB-2 but instead have proprietary MIBs, which the application may not support. There may be errors in the information presented on your network map for switches with proprietary MIBs. The following table contains information on switches, including those with proprietary MIBs, which the application does support. It also includes information on the known switches with proprietary MIBs that the application does not support. Refer to the application's web page www.flukenetworks.com/inspector for the most current information for known switch support.

Table 1. Switch Support Matrix (Yes indicates support for the identified switch)

See Also

The Agent's Device Discovery and Problem Reporting Processes

What Devices the Agent Discovers

Devices Missing from Your Network Map

Network Map Errors

The Agent's Device Discovery and Problem Reporting Processes

When you start the Agent, it begins discovering information about the devices that are part of the local broadcast domain that the PC that the Agent resides on is connected to. The Agent uses a combination of passive and active techniques to discover devices. By analyzing the information it gathers from the passive and active discovery tests, it can determine whether certain errors, warnings, and information have occurred. The Agent will then store the discovered information in its database.

Note

The Agent discovers most of the devices in a broadcast domain in about 10 minutes (on networks of 200 or fewer nodes). The SNMP analysis of the network will usually be complete in about 30 more minutes. The greater the number of switches and nodes on your network, the longer the SNMP analysis will take, requiring several hours for large networks.

Passive Discovery Tests

The Agent's passive discovery techniques include the following:

Listening to unicast traffic directed to the Agent's PC

Listening to broadcast and multicast traffic directed to the Agent's PC and to other devices in the Agent's broadcast domain

By listening to this traffic, the Agent can begin to identify the other devices in its sub-network.

Initial Active Discovery Tests

To determine what IP, IPX, and NetBIOS devices exist in its sub-network, the Agent does the following:

It sends a small set of broadcast messages.

Then, to the devices that respond, the Agent sends protocol-specific messages to determine if each device is local to the Agent's sub-network.

For each local IP device discovered, the Agent does the following:

It sends a DNS query to the DNS server to get the device's DNS name.

It sends an SNMP query to each IP device to get its SNMP System Group information. If a device runs an SNMP Agent and replies to the Agent's SNMP System Group query, the Agent then queries the device for its SNMP Interface Group information and for RMON information. For more information on how the Agent queries devices for SNMP and RMON information, refer to the SNMP Tab topic.

It sends additional queries to identify any IP services being offered by the device. (Refer to the IP Service Has Resumed on Device topic in the Console help for a list of possible IP services.)

For each local IPX device discovered, the Agent attempts to run userlist or nlist on the Novell server to determine the user login name for the device and to determine if the device is advertising any services. (Possible services include File service, Print service, Netware Directory service, Time Synchronization service, Net Manage service, Remote Access service, and RIP Router service.) It then uses this information to determine the best IPX name for the device.

For each local NetBIOS device discovered, the Agent sends additional queries to identify the device's domain/workgroup and to identify any services being offered by the device. (Possible services include Primary Domain Controller, Backup Domain Controller, and Master Browser.)

Note

The initial active discovery process can take up to 10 minutes on a small network (200 or fewer nodes). Therefore, it might be a while before all the discovered devices are shown in the Console's Explorer View.

Ongoing Active Tests

Once the initial active discovery tests are done, the following ongoing tests are performed:

The initial active discovery tests described above are repeated approximately every 90 minutes (or every 12 hours if you select Apply Large Network Discovery Algorithms in the Advanced Tab of the Agent).

For more information, refer to the Advanced Tab.

In addition, approximately every 2 minutes, the Agent queries each key device to determine whether it is active (that is, to determine its up/down status). An IP Ping (ICMP Echo) is used to query devices that use IP networking, and an IPX Ping (IPX Diagnostic) is used to query devices that use IPX networking. If a key device does not respond after multiple queries, a Key Device Not Responding error is reported in the Console's Problem Log.

In a similar manner, once every 90 minutes (or every 12 hours if you select Apply Large Network Discovery Algorithms in the Advanced Tab of the Agent) the Agent queries each non-key device to check its up/down status. However, to prevent the Console's Problem Log from being flooded with lots of status errors about non-key devices, no error is immediately reported if a non-key device does not respond. (Refer to the "Reporting of Errors, Warnings, and Information" section below for information on when and how a non-key device not responding is reported.)

Note

If you would like to have the up/down status of a particular non-key device checked and reported more often, you can make the device a key device.

Also, approximately every 2 minutes, the Agent queries all devices that are Utilization Sources to gather SNMP interface statistics. It uses this information to calculate the utilization percentage and error percentage for interfaces. The following error and warning conditions are reported in the Console's Problem Log (if the interface is being trended). For more information on trending, refer to Interface Trending Policy dialog box in the Console help.

Reporting of Errors (), Warnings (), and Information ()

The Agent can detect three categories of "problems": errors, warnings, and information.

Notes

For a list of all the errors, warnings, and information that can be detected by the Agent and reported in the Console's Problem Log and in the Problems tab in the Console's Device Properties dialog box, refer to Errors, Warnings, and Information that can be Discovered in the Console help.

From the Problems tab, you can control which types of errors, warnings, and information will be reported to the Console.

Once the Agent has performed its initial active discovery tests, it analyzes the information in its database to determine whether certain errors and warnings exist. If an error or warning exists, it is reported immediately to the Console. Therefore, you might see some errors and warnings soon after the Agent's initial active discovery tests are run and devices are shown in the Explorer View. For example, if the Agent determines that an IP device is using an IP address that makes it the only device in an IP subnet, the Agent reports an Incorrect IP Address error.

Note

The term "immediately," when used in the context of errors, warnings, and information being reported to the Console, means that the errors, warnings, and information will be displayed in the Console as soon as the display is refreshed. The display is refreshed automatically at the frequency specified in the Auto Refresh Frequency field of the Options dialog box . Additionally, you can force a refresh of the display at any time by pressing F5, using the Refresh item in the View menu, or the button on the Toolbar.

As the Agent performs its ongoing active tests and as it continues to gather information passively, it reports additional errors, warnings, and information as it detects them. For example, if, when the Agent queries a key device, the device does not respond, a Key Device Not Responding error is reported immediately. Therefore, you might see additional errors, warnings, and information as time goes on.

The one situation that is not reported as soon as it is detected is the Device No Longer Seen on Network change. No error is immediately reported if a non-key device does not respond to a query from the Agent. If this situation was immediately reported, the Console's Problem Log could be flooded with status errors when things like hosts being turned off occur. If a non-key device is still not responding to the Agent's queries after 72 hours, then a Device No Longer Seen on Network change is reported.

Other Things that Affect When You See Information Change in the Console

A few other things can affect when you see information change in the Console:

The automatic refresh rate selected in the Console's Options dialog box, which controls how often the Console automatically re-reads the entire database and re-displays the information in the Console.

Manually forcing a refresh of the Console's display, which also causes the Console to re-read the entire database and re-display the information in the Console.

Manually forcing the Agent to re-discover a device, which forces the Agent to re-discover and, if necessary, update the information in the database for the device currently being examined in the Console's Device Properties dialog box.

See Also

What Devices the Agent Discovers

Discovery of Devices with Proprietary MIBs

Index

A

About tab 13

Advanced Tab 11

Agent 30, 31, 32, 33

Configuring 18

Disabling 21

Seeing status of 24

Specifying startup mode 30

Starting 31

Stopping 34

Troubleshooting 14, 36

Agent: 14, 18, 21, 24, 34, 36

Alternate community string 9

SNMP 9

Alternate community strings 27

Appending information to the database  23

Applying changes 22

Automatic startup

specifying 30

B

Build number 13

Network Inspector application 13

C

Changes

making them take effect 22

Closing the Agent Manager's dialog box 17

Community strings 9, 27, 28

Configure

the E-mail feature 19

Configuring 10

E-mail notification 10

Pager notification 10

Configuring the Agent 18

Control What Problems Generate a Notification  45

Controlling what problems get reported in the Console  20

Copyright information 13

Network Inspector application 13

D

Database

Compaction 35

Using with standard database tools  35

Database

Appending information 23

Overwriting 23

Selecting a different 25

Database : 23, 25

Database Name field 25

Database/Address Tab 7

Database: 35

Default community string 28

Device Discovery and Reporting Processes  49

Device Properties

SNMP tab 9

Disabled radio button 21

Disabling the Agent 21

Discover E-Mail Servers 11

Enable/Disable 11

Discover HTTP Servers 11

Enable/Disable 11

Discovery of Devices with Proprietary MIBs  47

DNS server 29

specifying IP address 29

DSN 23, 25, 31

E

E-mail

configure 19

E-mail Configuration 10

Ensure That the Agent can 44

Ensure That the Agent Can Correctly Identify IPX Names  44

Error Messages 14, 36

Errors 26

thresholds 26

Event Log 14, 36

Event Viewer 14, 36

Events that can be logged to the Event Log  14

G

Gateway 29

Specifying IP address 29

General Discovery 12

Applying large network discovery algorithms  11

Global threshold levels 26

H

How this Help File is Organized 36

I

Introduction to the Agent Manager  3

IP address 29

Specifying the default router's 29

Specifying the DNS server's 29

IP Server Discovery 11

Enable/Disable 11

L

Large Network Discovery Algorithms  11, 12

applying 11

M

Making changes take effect 22

Manual startup

specifying 30

N

Names

Shown in SNMP tab of Device Properties  9

Network Inspector application

Build number 13

Serial number 13

Version number 13

Notification Tab 10

O

Overwrite check box 23

Overwriting the database 23

P

Problems

controlling which ones get reported in the Console  20

Problems Tab 20, 37

R

RMON community string 27

Router 29

specifying default IP address 29

S

Seeing the status of the Agent 24

Selecting a different database 25

Serial number 13

Identifying for the Network Inspector application  13

Service Dialog Box 30, 46

Service Tab 5

Set Global Threshold Levels 26

SNMP

Alternate community string 9

Community string 9

Contact 9

Description 9

Interface Group information 9

Location 9

Name 9

System Group information 9

Tab 9

SNMP community string fields 27, 28

SNMP Discovery 11

Enable/Disable 11

SNMP Tab 9

Specifying 27, 28, 29, 30

Alternate and RMON SNMP community strings  27

Default SNMP community string 28

Startup mode 30

The default router's IP address 29

The DNS server's IP address 29

Start the Agent Automatically When Windows Boots  30

Starting the Agent 30, 31

Automatically when Windows boots 30

Now 31

Startup button 30

Startup mode

specifying 30

Status field 24

Status messages 14, 24, 36

Status of the Agent

seeing 24

Stopping the Agent 34

System Log 14, 36

T

The Agent's Device Discovery and Problem Reporting Processes 49

The Agent's Device Discovery and Reporting Processes  49

Threshold 26

error and warnings 26

Threshold levels 26

Setting global 26

Thresholds

error and warnings 26

Tracing Enabled 31

ODBC Tracing warning 31

Troubleshooting messages 14, 36

U

Use the Event Viewer to View Agent Events  36

Using the database with standard database tools  35

Using the NT Event Viewer to view Agent events  36

V

Version number 13

Network Inspector application 13

W

Warning 31

ODBC Tracing is Enabled 31

Warnings 26

thresholds 26

What Devices the Agent Discovers  40