|
When you select Software Options from the Agent Configuration Utility main menu (option 18), special software options and their current status are displayed in another menu. Each option provides additional functionality in the SwitchProbe device. The device must have 512K flash to use the software option.
Table 11-1 lists the software options and the SwitchProbe devices that can use them.
You Can Use These Software Options... | On These SwitchProbe Devices |
---|---|
Resource Monitor | All SwitchProbes devices |
NetFlow Monitor | Ethernet |
VLAN Monitor | Fast Ethernet |
ART MIB support | All SwitchProbes devices |
Fast EtherChannel | Multiport Fast Ethernet |
WAN Decompression | Multiport T1/E1 WAN |
The following sections contain more information about using the licensed SwitchProbe software options:
The Resource Monitor option allows SwitchProbe devices to proactively and remotely monitor SNMP devices. As a dedicated proxy monitor that polls only at specific intervals, Resource Monitor can reduce the expensive and congestive regular polling required for traditional SNMP device management.
After you enable this option on a SwitchProbe device, the device can assess either SNMP get requests (for unwanted conditions) or IP ping requests to an IP device (to measure latency). The device temporarily stores the collected data until it is retrieved by the TrafficDirector application.
Resource Monitor supports the following types of responses from remote SNMP devices:
You can also set a TrafficDirector alarm on either type of Resource Monitor variable to ensure you are informed immediately of the condition or problem, or generate Trend Reporter reports.
Resource Monitor provides comprehensive monitoring without intensive ongoing communications to a remote SNMP device.
The IP ping request is a simple approach to test for device reachability. You can use Resource Monitor to:
Proxy SNMP monitoring also let you see current values of a specific MIB variable in real time or capture any changes in the value at periodic intervals that you specify (for example, log the value of the MIB variable every five minutes or once a day). You can use the logged data in a Trend Reporter report.
To generate the proxy SNMP status, the TrafficDirector application downloads MIB variables to an agent (that you either select from a list or from any that you add) to create proxy resources at the agent dedicated to track specific MIB status at a remote SNMP device.
To initiate the round trip delay assessment, the SwitchProbe device sends the IP ping request repeatedly based on an interval that you define, and calculates the average round trip of a series of ping attempts.
After you have enabled the Resource Monitor on the SwitchProbe device and set the tracking features for each participating agent using the TrafficDirector application, you can use several TrafficDirector features to monitor remote SNMP devices:
If you set an alarm so that a proxy SNMP MIB value or IP ping response exceeds or falls below a configured threshold value (for example, when disk space exceeds 90 percent or when the number of ping attempts exceeds five), an alert is automatically displayed in Alert Monitor.
You can also generate Trend Reporter reports to gauge behavior over timeuse the Network Response Time History report to reflect Round Trip Delay data, and the Proxy SNMP report for Proxy SNMP information.
For more information about the following topics, see Using the TrafficDirector Application:
NetFlow technology lets you configure the Cisco routers with the NetFlow option so you can redirect IP packet streams from all router ports to a preconfigured User Datagram Protocol (UDP) port and IP address. On those Cisco Systems routers with the NetFlow option enabled, you can use a SwitchProbe device for distributed IP monitoring to gather RMON, RMON2, and enhanced RMON information about the router backbone.
You can use the NetFlow Monitor option on Ethernet, Fast Ethernet, Multiport Fast Ethernet, and Gigabit Ethernet SwitchProbe devices.
In the Ethernet SwitchProbe device, the NetFlow Monitor option activates one special interface to receive the aggregated traffic of a Cisco Systems router.
In the Fast Ethernet, Multiport Fast Ethernet, and Gigabit Ethernet SwitchProbe devices, the NetFlow Monitor option lets you create the following interfaces:
When you install NetFlow Monitor in a SwitchProbe device, the device creates a special internal interface (numbered 50) that you can specify as the UDP port. You can monitor NetFlow traffic by adding interface 50 to the TrafficDirector application as an agent. You can configure agent groups to support integrated displays for multiple routers, and use the TrafficDirector Host and Conversation Lists feature to monitor backbone loading and other interactions.
To monitor NetFlow traffic, you must configure the SwitchProbe device with the following information:
After you enable the NetFlow Monitor software option on a SwitchProbe device, you must set additional parameters for the following reasons:
To configure the device to receive aggregate traffic through special virtual interface 50, follow these steps:
Step 1 To access command-line mode, enter 11 and press Enter.
Step 2 To specify the NetFlow UDP port number, enter the following command:
set netflow_port <port number>
where <port number> is Cisco router NetFlow port number. The default is 9991.
Step 3 To specify the UDP port speed, enter the following command:
set netflow_speed <port speed>
where <port speed> is the sum of all router interface speeds (in bps). For example, if the router has 10 interfaces each at 10 Mbps, you should specify 100Mbps (100,000,000 bps) as the NetFlow port speed.
Step 4 To exit command-line mode, enter QUIT and press Enter.
The Agent Configuration Utility main menu is displayed again.
The enabled software options do not take effect until you reboot or reset the SwitchProbe agent. To do so, see "Resetting a SwitchProbe Agent" in "Configuring SwitchProbe Devices."
To configure a Fast Ethernet SwitchProbe device to receive NetFlow packets of a maximum of eight individual router interfaces on eight separate special virtual interfaces in the device, follow these steps:
Step 1 To access command-line mode, enter 11 and press Enter.
Step 2 Enter the following command for the packets on each router interface that you want to direct to the SwitchProbe device.
set netflow_ifn add <netflow_ifn> <netflow_speed>
<router _ip_address> <router_ifn>
where:
If you enter 0 as a router interface number, the SwitchProbe device interprets that number as an unspecified router interface and directs the aggregate of NetFlow packets from the specified router (router_ip_address) to the specified special virtual interface of the SwitchProbe device (netflow_ifn).
For example, to direct NetFlow packets from interface 2187 of a router (with an IP address of 100.100.210.43) to special virtual interface 57, you would use this command:
set netflow_ifn add 57 9600 100.100.210.43 2187
Step 3 To exit command-line mode, enter QUIT and press Enter.
The Agent Configuration Utility main menu is displayed again.
The enabled software options do not take effect until you reboot or reset the SwitchProbe agent. To do so, see "Resetting a SwitchProbe Agent" in "Configuring SwitchProbe Devices."
You use the TrafficDirector application to define the agents to monitor these special NetFlow Monitor interfaces from the SwitchProbe device. You can also configure agent groups to support integrated displays from multiple routers in real-time monitoring applications. Then, using the TrafficDirector Domain Editor application, you can create a generic IP domain to isolate and track NetFlow traffic using AS number-mapping conventions to aggregate the data for billing and chargeback.
You can also generate Trend Reporter reports to gauge behavior over timeuse the Router Backbone Usage report to reflect NetFlow data. For more information about configuring agents to track NetFlow data, or generating relevant reports in Trend Reporter, see
Using the TrafficDirector Application.
The TrafficDirector application can recognize data collected from special NetFlow interfaces configured on SwitchProbe devices from either:
To configure the TrafficDirector application to interpret aggregate traffic through special interface 50, see Using the TrafficDirector Application.
After you set up an agent for each special interface, the TrafficDirector application can monitor the traffic individually associated with each interface. NetFlow special interfaces are automatically learned after you define the device IP address and click View in Configuration Manager.
When you install properties files on a multi-port SwitchProbe device, the same file is installed across all special interfaces whenever you define these interfaces as individual agents in Configuration Manager.
To track V. 5.0 NetFlow data using AS numbering conventions, you should create a custom domain and add it to a shared properties file that you create.
You must properly define the NetFlow export statement on the routers and switches to transfer the AS form of data.
To capture traffic using AS numbering conventions, follow the steps to create a domain as outlined in Using the TrafficDirector Application. A summary of this process is provided here.
Step 1 Use the Domain Editor application to define a generic domain.
Step 2 Name the domain, such as ASDOM. The host address mode should be SUBNET, Domain Type Inclusive, Packet Type All and Selected Filters IP.
Step 3 Create a new shared properties file.
For the configuration to take effect, you must install the relevant properties file on the agent using Configuration Manager.
When retrieving NetFlow data for the generic domain ASDOM, expect to see the following:
The VLAN Monitor allows Fast Ethernet, Multiport Fast Ethernet, and Gigabit Ethernet SwitchProbe devices to detect RMON statistics for each Virtual Local Area Network (VLAN) on a network link between two switches or on a port in a switch. VLAN Monitor isolates the activity of each VLAN (as a separate virtual interface), so you can monitor high-speed interswitch trunks and critical links. VLANs are carried between two switches over Fast Ethernet links using either ISL or 802.lq trunking protocols.
By enabling this mode in a SwitchProbe device, you can monitor packets using VLAN IDs instead of MAC addresses. VLAN Monitor provides additional priority statistics for any VLAN that supports the SMON-MIB standard.
VLANs let you build software LANs regardless of each user's physical location. VLANs imply a broadcast domainsimilar to a subnetwithin which broadcasts are communicated to all hosts. Routers are required for inter-VLAN communication.
The VLAN Monitor option directs the selected Fast Ethernet, Multiport Fast Ethernet, or Gigabit Ethernet SwitchProbe interface to read the protocol headers and determine the VLAN ID number of each packet that passes through the interface.
Each incoming Ethernet ISL packet is reassembled (the MAC address of the packet is eliminated and the Ethernet data in the incoming packet is appended to the reassembled packet).
Similarly, each incoming Token Ring and FDDI packet is treated in the same manner. As a result, the VLAN Monitor can collect RMON2 traffic (including hosts and conversations by VLAN ID) of the monitored segment.
If the VLAN Monitor option is disabled, the Ethernet ISL packet is reassembled again as a simple Ethernet packet. This also happens with packets generated on Token Ring and FDDI LANs.
Using a SwitchProbe device, the TrafficDirector application can recognize data collected from network segments that include VLAN-based data or from VLAN data on interswitch links. After you set up an agent for each virtual interface, you can use the TrafficDirector application to monitor the traffic individually associated with the interface (by VLAN ID).
You can also define an agent group to track the separate virtual interfaces collectively. An agent group is a convenient way to monitor multiple segments simultaneously, or update and install properties files across a number of agents simultaneously.
Virtual interfaces are automatically learned by the TrafficDirector application after you define the SwitchProbe IP address and enter other relevant configuration information in Configuration Manager.
You use a shared properties file to identify the domains you wish to track on each virtual interface. You can also create a custom properties file for individual virtual interfaces so you can set specific logging parameters and alarms for the interface.
Logging parameters define how often the device is polled for data; alarms define whether an alert is sent to the TrafficDirector console every time a user-defined threshold is exceeded. This custom file cannot be shared with other virtual interfaces.
You can monitor data from individual VLANs when you use many Cisco Catalyst family switches are attached to a SwitchProbe device for dedicated monitoring. You can view data by VLAN ID on any switch port or trunk (that conforms to ISL or 802.lq trunking protocols) between VLANs.
If your switch also supports the SMON standard, you can use the VLAN Monitor application with the TrafficDirector application to view data by:
You can also generate Trend Reporter reports to gauge behavior over timeuse the VLAN Usage report to reflect VLAN priority data.
For more information about the following topics, see Using the TrafficDirector Application:
You use the TrafficDirector Property Editor application to create properties files that you can install on a SwitchProbe device for virtual interfaces that track VLAN information.
For virtual interfaces that require unique logging or alarm capabilities, you can create a custom properties file. In this case, you create the properties file for one virtual interface at a time, and it cannot be used by other virtual interfaces on the same SwitchProbe device. After they are installed, the settings in the custom properties file overrides any shared properties file in effect.
For more information about creating shared and custom properties files for agents and how they apply to VLAN-based virtual interfaces, see Using the TrafficDirector Application.
The ART MIB is an extension to the RMON2 standard that lets SwitchProbe devices collect a series of application response time (ART) statistics, such as connection retries, traffic load, and response time buckets. To collect these types of statistics with the TrafficDirector ART Monitor application, you must enable this option in the SwitchProbe device.
The ART MIB option measures how network performance supports the flows of specific applications. The most important characteristic is a request-response structure within the applications, such as the ongoing acknowledgment of data transferred within TCP conversations.
You can use the ART MIB to identify request-response time exchanges between a server and clients for a specified application layer domain. Using ART Monitor, you can determine which applications are overloading the network or not responding as expected.
This feature lets an ART MIB-enabled agent determine the associated response time by matching up two packets. When a packet is seen in a conversation, its transmission time is noted. The corresponding response is seen coming back (typically from the server) and the time is noted. The time difference between the request and response is measured in milliseconds; the elapsed time is a round-trip time that includes some processing time.
There is a user-configurable report interval in ART MIB agents for determining when the data is presented and retrieved from the SwitchProbe device. The default ART MIB reporting interval is 15 minutes.
Using a 15-minute report interval as an example, the ART MIB agent makes response time measurements and computes an average at the end of this time period. When the 15-minute interval ends, the agent memory tables are updated (older information is replaced).
For the next 15-minute interval, that data can be retrieved at any time. When another 15-minute-interval has elapsed, the tables are updated again. In this way, information is summarized and maintained as stable and accessible for retrieval, so that the agent and the TrafficDirector application need not be tightly synchronized.
When you select Change ART MIB Parameters from the Agent Configuration Utility main menu (option 21), several options are displayed in another menu. These options let you define time, slice, and report values that the SwitchProbe device uses to prepare application response statistics.
The first six options of the ART MIB Parameter let you modify the time ranges into which ART statistics are grouped. Each option uses the value of the preceding options as the minimum, and its own value as the maximum. For example, if an option value is 72 (ms) and its preceding option value is 30 (ms), the range of the option includes responses that take more than 30 ms but less than or equal to 72 ms.
Tables 11-2 and 11-3 describe the ART MIB parameters.
This Option... | Displays and Lets You Modify the Maximum Boundary (in ms) of... | Maximum Boundary is Less Than Value of... | Minimum Boundary is Greater Than Value of |
---|---|---|---|
Change art_resp_time1 | An ART time range | Change art_resp_time2 | 0 |
Change art_resp_time2 | A second ART time range | Change art_resp_time3 | Change art_resp_time1 |
Change art_resp_time3 | A third ART time range | Change art_resp_time4 | Change art_resp_time2 |
Change art_resp_time4 | A fourth ART time range | Change art_resp_time5 | Change art_resp_time3 |
Change art_resp_time5 | A fifth ART range | Change art_resp_time6 | Change art_resp_time4 |
Change art_resp_time6 | A sixth ART range | Change art_resp_timeout | Change art_resp_time5 |
Change art_resp_timeout | A final ART range | | Change art_resp_time6 |
Note This value establishes a timeout time limit. When that limit is reached (no responses are received within a specified period of time), the device empties the ART counters. |
This Option... | Lets You |
---|---|
Change art_report_interval
| Display and modify the frequency (in seconds) at which the contents of a table of ART statistics (updated one interval ago) is replaced with a table of the most recent ART statistics; the ART counters are also reset to 0. For example, if you enter 900 in this field, the contents of a table of ART statistics is replaced every 15 minutes with a new table of the most recent ART statistics, and the ART counters are reset to 0. |
Change art_report_size | Display and modify the maximum number of client/server pairs for which you want to collect ART statistics. |
Change art_aging_frequency | Lets you specify the frequency at which the ART MIB tables are replaced, if they are full. |
Toggle start_art | Determine if the device tracks ART statistics. If set to off, the device does not collect ART statistics; therefore, the TrafficDirector ART Monitor application will have no data to present. |
To modify the value of an ART MIB parameter, follow these steps:
Step 1 From the Change ART MIB Parameters menu, enter a parameter option number and press Enter.
Step 2 Enter the desired value and press Enter.
The new value is displayed next to the parameter.
Step 3 To return to the Agent Configuration Utility main menu, enter 12 and press Enter.
The Agent Configuration Utility main menu is displayed again. The new values do not take effect until you reboot or reset the SwitchProbe agent. To do so, see "Resetting a SwitchProbe Agent" in "Configuring SwitchProbe Devices."
The TrafficDirector Response Time (ART) Monitor displays statistics collected by an agent supporting the ART MIB.
You can use ART Monitor for:
For more information about the ART Monitor application, see Using the TrafficDirector Application.
Table 11-4 describes the TrafficDirector Trend Reporter reports you can use to monitor ART performance.
This Report... | Provides This Information |
---|---|
Application Response Time History | Tracks ART client and server statistics for traffic on one segment for discrete intervals during the reporting period. Examines detailed data samples for ART client and server statistics. |
Application Response Time Server History | Tracks ART server statistics averaged from a selected number of clients for traffic on one segment for discrete intervals during the reporting period. Examines detailed data samples for ART server statistics. |
Application Response Time Summary | Tracks and accumulates ART client and server statistics for traffic on one segment during the reporting period. Examines summarized data for ART client and server statistics. |
Application Response Time Server Summary | Tracks and accumulates ART server statistics averaged from a selected number of clients for traffic on one segment during the reporting period. Examines summarized data for ART server statistics. |
The Fast EtherChannel software option lets you aggregate the traffic of two or four separate full-duplex 200 Mbps Fast Ethernet segments into one full-duplex trunk.
The Fast EtherChannel (FEC) option only functions on four-port Multiport Fast Ethernet SwitchProbe devices when you configure all Fast Ethernet ports on the device to full-duplex mode, and two full-duplex segments populate the four ports on the SwitchProbe device.
For information about configuring the SwitchProbe mode, see "Selecting the Probe Mode (Multiport Fast Ethernet and Gigabit Ethernet Devices)" in "Configuring SwitchProbe Devices."
For more information about defining agents for FECs, see Using the TrafficDirector Application.
The WAN Decompression licensed software option lets Multiport T1/E1 WAN SwitchProbe devices monitor a WAN link that contains compressed data.This is a hardware-based compressionthe device internally decompresses the WAN traffic being monitored.
WAN decompression currently supports only these compression types:
To enable a software option, follow these steps:
Step 1 Access the Agent Configuration Utility main menu.
Step 2 Enter 18 and press Enter.
Step 3 Enter the option number and press Enter.
The option is activated in the SwitchProbe device. Some software options require that you configure special parameters as described in Table 11-5 .
Software Option | Special Parameter Configuration |
---|---|
Resource Monitor | No special parameters. Proceed to Step 4. |
NetFlow Monitor | |
VLAN Monitor | No special parameters. Proceed to Step 4. |
ART MIB Monitor | See "Using ART MIB Parameters." |
Fast EtherChannel | No special parameters. Proceed to Step 4. |
WAN Decompression | No special parameters. Proceed to Step 4. |
Step 4 To return to the Agent Configuration Utility main menu, enter 32 and press Enter.
The Agent Configuration Utility main menu is displayed again.
The enabled software option does not take effect until you reboot or reset the SwitchProbe agent. To do so, see "Resetting a SwitchProbe Agent" in "Configuring SwitchProbe Devices."
The enabled software option and any related settings remain installed in the SwitchProbe device until you disable it.
To disable a software option, follow these steps:
Step 1 From the Software Options menu, enter the option number and press Enter.
The option is disabled in the SwitchProbe device the next time you reboot or reset the device, and remains disabled until you enable it.
Step 2 To return to the Agent Configuration Utility main menu, enter 32 and press Enter.
Posted: Wed Oct 2 08:10:28 PDT 2002
All contents are Copyright © 1992--2002 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.