User Guide for Cisco Unified Operations Manager 2.2
Using Performance Graphs
Downloads: This chapterpdf (PDF - 264.0KB) The complete bookPDF (PDF - 13.05MB) | Feedback

Using Performance Graphs

Table Of Contents

Using Performance Graphs

How to Use Performance Graphs

What Metrics Can I Include on a Graph?

Performance Graphing Notes

Launching a Performance Graph

Working with Graphs

Understanding Graphs and Getting More Information

Working with a Merged Graph

Troubleshooting Performance Graphs


Using Performance Graphs


These topics explain how to use performance graphs:

How to Use Performance Graphs

Working with Graphs

How to Use Performance Graphs

Cisco Unified Operations Manager (Operations Manager) allows you to select and examine changes in network performance metrics. You can select, display, and chart network performance data in real time. The performance graphs are accessed through the Service Level View, Alert Details page, and Node-to-Node Tests page.

You can create performance graphs from the data that is collected when:

Voice utilization polling is enabled for devices. (See Editing Polling Parameters, page 19-13.)


Note Voice utilization polling is disabled by default.


Data is available on disk from node-to-node tests that you have configured. (See Working with Node-To-Node Tests, page 11-1.)

The following topics describe the data you can graph and help you to understand the information displayed:

What Metrics Can I Include on a Graph?

Performance Graphing Notes

Launching a Performance Graph

What Metrics Can I Include on a Graph?

When you create a performance graph (see Launching a Performance Graph), you select devices and then select among metrics that are appropriate to those devices. Table 7-1 summarizes the types of metrics that you can graph for each device type.

Table 7-2 summarizes the node-to-node test metrics that you can graph.

Table 7-1 Utilization Metrics that You Can Graph by Device Type 

Device Type
CPU
Memory
Calls
Port Utilization
Channel Utilization
Device-Specific Usage

Cisco Unified Communications Manager

X

X

X1

 

BRI

FXS

FXO

T1/E1 PRI

T1 CAS

Cisco Unified Communications Manager Resource Utilization:

MOH multicast

MOH unicast

MTP resource

Transcoder

Hardware conference

Software conference

Percentage conference active

Percentage conference streams active

Location bandwidth available

CTI links active

Registered analog access

Registered MGCP gateways

Registered hardware phones

Cisco Unified Communications Manager Express

X

X

X2

   

IP phones registered

Key IP phones registered

Cisco Unified Communications Manager-controlled MGCP3 gateways

     

FXS

FXO

T1/E1 PRI

T1 CAS

BRI4

 

Cisco Unified Communications Manager or Cluster

           

Voice Mail Gateway

X

   

Voice mail and PBX ports

   

Gatekeeper

X

X

     

Zones5

Cisco IOS gateways

X

X

X6

E&M

FXO

FXS

T1/E1 PRI

T1/E1 CAS7

BRI

DSP8

SRST

X

X

     

Minutes in SRST mode

Unity

X

X

 

Inbound

Outbound

   

Unity Express

X

       

Capacity utilization

Session utilization

Orphaned mailboxes

Unity Connection

X

X

 

Inbound

Outbound

Total number of active ports

Total number of ports

Total number of active inbound ports

Total number of inbound ports

Total number of active outbound ports

Total number of outbound ports

   

Unified CCE

X

X

     

Number of contact center agents currently logged into Unified CCE

Number of calls in progress in Unified CCE

Number of inbound calls per second

1 Active calls on Cisco Unified Communications Manager.

2 E-phone active call legs.

3 MGCP = Media Gateway Control Protocol.

4 BRI channel status for MGCP gateways is supported only with Cisco Unified Communications Manager 4.1 and later.

5 Zone information includes bandwidth, requests, and errors.

6 Active call legs on Cisco IOS gateway.

7 T1/E1 CAS channel status for Cisco IOS gateways is only reported for Cisco AS5200 Series Universal Access Servers.

8 DSP = Digital signal processor.


The data files for the performance metrics listed in Table 7-1 are located on the server, in the NMSROOT\data\gsu\_#GSUdata#_ directory. If you do not have access to the directory, contact a local administrator for the server where Operations Manager is installed. The filenames are created using the device name and the date. These files are kept for 72 hours, after which they are purged.


Note NMSROOT is the directory where Operations Manager is installed on your system. If you selected the default directory during installation, it can be entered as "C:\Program Files\CSCOpx" or C:\PROGRA~1\CSCOpx.


Table 7-2 Graphable Metrics for Node-to-Node Tests

Node-to-Node Test
Graphable Metrics

UDP Jitter for VoIP

Source-to-destination packet loss.

Destination-to-source packet loss.

Source to destination jitter.

Destination to source jitter.

Average latency.

Node-to-Node quality.

Ping Echo

Round-trip response time.

Ping Path Echo

Round-trip response time.

UDP Echo

Round-trip response time.

Gatekeeper Registration Delay

Registration response time.


The data files for the performance metrics listed in Table 7-2 are located on the server, in the NMSROOT\data\N2Ntests directory. The data files in this folder are purged after 31 days. If you do not have access to the directory, contact a local administrator for the server where Operations Manager is installed.


Note NMSROOT is the directory where Operations Manager is installed on your system. If you selected the default directory during installation, it can be entered as "C:\Program Files\CSCOpx" or C:\PROGRA~1\CSCOpx.


Performance Graphing Notes

This section contains information you should be aware of when working with performance graphs.

Table 7-3 Performance Graphing Notes 

Summary
Explanation

An MGCP gateway on a Catalyst 6000 switch. When you have all three capabilities (voice gateway, switch, and MGCP) performance graphing cannot graph all the data. Only the common metrics are available for graphing.

When graphing performance metrics for a device that has these three capabilities (voice gateway, switch, and MGCP) you will only be able to graph the common metrics.

1. Select the device for which you want to graph performance metrics.

2. Select Performance. A metrics dialog box appears.

3. Select the desired metrics, and click Next. A second dialog box appears, listing the MGCP gateway ports.

4. Select a port, and then click View.

Note In the Alert Details page you will not have the option to graph HighUtilization events.

A voice gateway, MGCP, and H323 on a router. When you have all these capabilities on one device, each metric displays two graphs.

When graphing performance metrics for a device that has these capabilities (voice gateway, MGCP, H323, and router), each metric displays two graphs.

Also, when graphing multiple devices or devices that have multiple polling intervals, the least common multiple is used to plot the x axis.

Real-time graphs will refresh at this common polling interval.

Location Bandwidth Available text box. The location bandwidth is located on the Cisco Unified Communications Manager system.

Enter the location on the Cisco Unified Communications Manager system. This location is configured through Cisco Unified Communications Manager. If you have questions regarding the location bandwidth, see your Cisco Unified Communications Manager documentation.

1. Select the device for which you want to graph the location bandwidth availability.

2. Select Performance. A metrics dialog box appears.

3. Select the Location Bandwidth Available check box.

4. Enter the location on the Cisco Unified Communications Manager system where the location bandwidth resides.

5. Click View.

Zone Home text box. The zone is located on the gatekeeper system.

Enter the location on the gatekeeper system. This location is configured through the gatekeeper. If you have questions regarding the location bandwidth, see your gatekeeper documentation.

1. Select the device for which you want to graph the zone home.

2. Select Performance. A metrics dialog box appears.

3. Select the Zone Home check box.

4. Enter the location on the gatekeeper where the zone home resides.

5. Click View.

In the Select Metrics dialog box for Unified CCE, you will see the following fields:

Agents logged on

Calls in progress

Inbound Calls per second

To view a performance graph for any of these metrics, you must enter the name of the Unified CCE instance for which you want the information, in the field next to Instance Name.

Each Unified CCE contains a list of enterprise contact center applications, which are identified by their instance name.

Operations Manager will not display a performance graph for all Unified Contact Center Enterprise (CCE) machines.

Operations Manager only collects data for Unified CCEs that have router capabilities. To verify your Unified CCE's capabilities, you can run a device report through device management, and look for Unified CCE or IPCC Router in the Device Capabilities column of the Device report.

You cannot view a performance graph for Unified CCEs that do not have router capabilities. Instead, you will get a message window stating such.


Launching a Performance Graph

The performance graphs are available through the following:

Service Level View—See Launching Operations Manager Tools from the Service Level View, page 2-20.

Route List and Route Group Report—See Viewing the Route List and Route Group Report, page 2-25.

Alert Details page—See Understanding the Layout of the Alert Details Page, page 3-10.

Detailed Device View—See Understanding the Layout of the Detailed Device View, page 3-19.

Node-to-Node Tests page—See Viewing Test Trending, page 11-16.

Before you Begin

Verify that Operations Manager is monitoring the devices for which you want to collect utilization statistics. This includes the Cisco Unified Communications Manager that the ports are registered to. See Verifying Device Import, page 16-23.

Enable the voice utilization polling settings. By default the voice utilization polling settings are not enabled. Operations Manager uses the statistics gathered during voice utilization polling for charting network performance. See Voice Utilization Settings—Polling, page 19-25 and Editing Polling Parameters, page 19-13.

Review the Performance Graphing Notes.

Use this procedure to launch a performance graph from the Service Level View.


Step 1 In the Service Level View (see Using the Service Level View, page 2-1), right-click on the device for which you want to see the performance metrics.


Note If you want to select multiple devices, hold the Ctrl key when clicking the devices in the map view.


Step 2 From the menu select Performance. The Select Metrics dialog box appears.

Step 3 Select one or more desired metrics, and click View Graph. A performance graph window appears, displaying one or more graphs (see Figure 7-1).


Note If not all expected values are plotted on a graph, the most likely reason is that one or more values are very small in comparison to the maximum value. For example, for the values 250, 2, and 1, the smaller values, 2 and 1, will not be plotted.



Note If you receive errors while creating performance graphs, you can look in the Error.log file for error details. The log file is located in the NMSROOT\data\gsu\_#GSUdata#_ directory. Also, see Troubleshooting Performance Graphs.



Working with Graphs

Performance graphs provide real-time information and historical information. When you launch a performance graph, one line graph is displayed for each metric that you select. Each line graph contains 16 data points displayed in real time. All the performance graph windows have the same layout as shown in Figure 7-1.

Figure 7-1 Performance Graph Window Layout

1

Showing Lists. See Showing Lists.

3

Graph display pane.

2

Show All and Merge radio buttons. See Show All or Merge.

   

Showing Lists

By default, these lists display:

Real Time—Select the number of hours of data to graph, up to a maximum of 72 hours.


Note If you select a device that does not have data available for the selected time interval, a message appears stating such. An empty graph appears and automatically refreshes periodically. At any time, you can change the time interval to get historical information, if there is any.


Line Graphs—Select Bar Chart or Area Chart.

With 16 data points—Select up to a maximum of 240 data points.

Show All or Merge

You can show all graphs or merge all graphs regardless of the unit of measure. When you select the Merge radio button, the merged graph is scaled to show all the metrics in the single graph. Also, a new Showing list appears, displaying Scaled. You can select Default in place of Scaled from the list. For an example of a merged graph, see Working with a Merged Graph.

Understanding Graphs and Getting More Information

Figure 7-2 shows two graphs, each with the following information:

Title—Reflects the metric.

X axis—Shows a time line.

Y axis—Displays the unit of measure.

Legend—Includes the name of the device from which the data was gathered and the name of the metric.

Figure 7-2 Default Display of a Performance Graph Window

The BRI Channel Utilization graph, on the right in Figure 7-1, includes a gray line at 0.00 percent. There is no BRI channel utilization data for the device. The gray message box that is shown appears when you hover your cursor over a gray line or gray area in a graph. If you click the gray message box and more information is available, a dialog box will appear.

Working with a Merged Graph

When you select the Merge radio button, individual graphs are merged into a single, scaled graph as shown in Figure 7-3.

Figure 7-3 Merge Scaled Performance Graph

In a merged, scaled graph, the following information is displayed:

X axis—Shows a time line.

Y axis—Shows numbers, but no unit of measure. The unit of measure is included in the legend.

Legend—Includes the name of the device from which the data was gathered, the name of the metric, the ratio by which the data is scaled, and the unit of measure.

To remove scaling from a merged graph, in place of Scaled, select Default. When you select Default, the legend includes only the device name and the metric.

Troubleshooting Performance Graphs


Note For information about what is displayed on graphs, see Understanding Graphs and Getting More Information.


This section contains information that will help you if you encounter problems generating performance graphs.

If you encounter an error, it will likely appear either when you select Performance Graphing from the menu, or when Operations Manager is checking for the data file to graph.

In the first case (when selecting Performance Graphing), you will see an error message that describes the problem and an action to take.

In the second case (when Operations Manager is checking for the data file), an error message saying something like No data file is available appears. The error messages appear in the performance graphing log file (Error.log) located at NMSROOT\data\gsu\_#GSUdata#_ directory.


Note NMSROOT is the directory where Operations Manager is installed on your system. If you selected the default directory during installation, it can be entered as "C:\Program Files\CSCOpx" or C:\PROGRA~1\CSCOpx.


Table 7-4 describes the errors and their possible causes, for both of these types of situations. Possible causes for the errors provide a starting point for you to investigate and take corrective action.

Table 7-4 Troubleshooting Performance Graphing Errors 

Error
Possible Causes

Cannot collect data.

Account and credentials are not the same on all Cisco Unified Communications Managers in the cluster. If this is required, see Determining the Media Server Account to Use for Cisco Unified Communications Manager Access, page 16-42.

HTTP server problems:

HTTP server on the device is down.

HTTP server is operational, but the Cisco Unified Communications Manager is down.

Device unreachable due to a network problem.

Performance Monitor process on the media server is down.

The Cisco Unified Communications Manager that the MGCP gateway is registered to is not in Operations Manager inventory.

Device capability is not supported. (Performance graphing supports the following: Cisco Unity, Cisco Unity Express, Cisco Unified Communications Manager, Cisco Unified Communications Manager Express, SRST, H.323 devices, and Voice Mail Gateways.)

Device is suspended or deleted.

Device platform is not support.

For device support information, see Supported Device Table for Cisco Unified Operations Manager on Cisco.com at http://www.cisco.com/en/US/products/ps6535/products_device_support_tables_list.html.

Cannot collect data because of the following:

The username or password for the device is empty.

The system has the wrong credentials for the device.

The device does not have credential information.

No credentials in Operations Manager.

Incorrect credentials in Operations Manager.

Note To add credentials, see Editing Device Configuration and Credentials, page 16-32.

Cannot collect data from the device because of the following:

A processing error occurred.

Parsing or processing errors occurred.

Internal initialization errors occurred.

Initialization problems occurred in the device data collector.

Incorrect Cisco Unified Communications Manager version. Check the following:

The version of the Cisco Unified Communications Manager that is running on the device.

The Cisco Unified Communications Manager version number that is stored in Operations Manager. For instructions, see Viewing Device Details, page 16-34.

Note If the Cisco Unified Communications Manager version number that is stored in Operations Manager is incorrect, re-add the device. See Understanding the Device and Credentials Repository, page 16-4.

For device support information, see Supported Device Table for Cisco Unified Operations Manager on Cisco.com at http://www.cisco.com/en/US/products/ps6535/products_device_support_tables_list.html.

Cannot collect WMI counters from the device.

The device maybe unreachable due to network problems.

An incorrect hostname, username, and/or password was used.

The Remote Procedure Call (RPC) service is not running on the device.

The Current user does not have permission to query WMI.

Cannot collect data from the device. The certificate hostname/IP Address cannot be mapped to the URL hostname/IP Address.

The device is not in DNS.

Incomplete data collected because an error occurred in communicating with the device.

Incorrect Cisco Unified Communications Manager version. Check the following:

The version of the Cisco Unified Communications Manager that is running on the device.

The Cisco Unified Communications Manager version number that is stored in Operations Manager. For instructions, see Viewing Device Details, page 16-34.

Note If the Cisco Unified Communications Manager version number that is stored in Operations Manager is incorrect, re-add the device. See Understanding the Device and Credentials Repository, page 16-4.

For device support information, see Supported Device Table for Cisco Unified Operations Manager on Cisco.com at http://www.cisco.com/en/US/products/ps6535/products_device_support_tables_list.html.

Cannot collect data because of the following:

The device returned no data from a required MIB.

The device received no MIB data.

No data from a required MIB.

A required MIB is not populated on the device.

No MIBs returned data.

Device is unreachable due to a network problem.

Device credentials do not contain a valid SNMP community read string.

SNMP response slow; data collection timed out.

The rate of queries on the Cisco Unified Communications Manager exceeds the limit.

An error has occurred in the data processing stage.

Too many queries on a Cisco Unified Communications Manager 4.0 or later.

Note Check the polling settings (see Viewing Polling Parameters, page 19-12); they should not be less than three minutes.

The Cisco Unified Communications Manager did not have enough time to handle the query requests.

An error has occurred in the data processing stage.

Query exceeded time limit on Cisco Unified Communications Manager 4.0 or later.



Note When working with performance graphs, remember the following:

If you are not able to collect performance data and you do not see an error message (either a popup message or a message in the log file) indicating the problem, you should verify the status of the device. To do so, use the View/Rediscover/Delete Devices page (see Verifying Device Import, page 16-23). If the device is in the Unreachable state, verify that the device's credentials are correct and rediscover the device (see Performing Manual Inventory Collection on Devices, page 16-33).

If a gray line or a gray area appears in a graph, hover your mouse over it to obtain a tooltip with an explanation.

To collect performance data for Cisco Unity Connection, Cisco Unity, or Cisco IP Contact Center, the Windows Management Instrumentation (WMI) credential is required. When adding these devices to Operations Manager, verify that the WMI username and password are provided.