Cisco Active Network Abstraction Administrator Guide, 3.6.5
System Health Diagnostics
Download this chapter
System Health DiagnosticsSystem Health Diagnostics
Download the complete book
PDF of Cisco ANA 3.6.5 Administrator GuidePDF of Cisco ANA 3.6.5 Administrator Guide (PDF - 3 MB)
give_us_feedback

Table Of Contents

System Health and Diagnostics

Introduction

Getting Started

Toolbar

Graphs

Graph Types

Web Page Options

MC Loads Page

MC server-ip Load Page

Invoking Additional Parameters

Use Cases

Use Case 1

Use Case 2


System Health and Diagnostics

This chapter describes how to work with the system health and diagnostics tool, and the various aspects of the Cisco ANA system that can be monitored.

This chapter includes the following sections:

•Introduction

•Getting Started

•Invoking Additional Parameters

•Use Cases

Introduction

The system health and diagnostics tool provides system resource utilization information for the gateway and units (physical, allocated, and used). It monitors various aspects of the Cisco ANA system, such as the size of the Java heap or AVM CPU usage, and enables you to verify that the gateway, units, and AVMs are functioning correctly.

Getting Started

This section provides instructions for accessing the tool. The tool is password-protected to ensure security. Before you start working with the tool, make sure you have your username, password, and the Cisco ANA gateway IP address or host name that you require.

Note The connection to the tool is via the HTTPS protocol and authentication method.

To access the tool:

Step 1 Enter https://gateway_ip:1311/graphs in your browser where gateway_ip is the gateway IP address.

A security alert is displayed regarding the site certificate.

Step 2 Click Yes to proceed.

Step 3 When prompted, enter the username and password.

Note The username and password were defined during the system installation as User name and Password for the web monitoring tool. The username entered here is not the same as the one used in Cisco ANA applications.

Note To change the username and password, log into the gateway as user sheer and run the following command:
$ SHEERHOME/utils/apache/bin/htpasswd -cb $SHEERHOME/Main/webroot/.passwd new-username new-password

The MC Loads page (Units monitoring page) is displayed. The MC Loads page displays a combined graph of the loads for the gateway and units in a graphical representation.

Figure 12-1 shows the MC Loads page (Units monitoring page).

Figure 12-1 MC Loads Page

1

Details

2

Toolbar

3

Web page options

4

IP Hyperlink

5

Units row

6

Gateway row

7

Graph types in columns

 

The web pages that are displayed in the tool can have the following components:

•Details of the IP address of the gateway or unit and the current date and time.

•Toolbar.

•Web page options.

•IP hyperlinks—Clicking the hyperlinked IP address of a gateway or unit displays the AVMs for the selected gateway or unit.

•Table of graphs—Each horizontal row represents a gateway or a unit, and each column represents a different graph type.

Toolbar

The toolbar enables you to adjust the graph period and columns that are displayed (see Figure 12-2).

Figure 12-2 Toolbar

The following drop-down lists appear in the toolbar:

•Period—Enables you to set the period of time for the graphs in the page.

•Add column or Remove column—Enables you to add or remove graph types from the page.

Step 1 Choose an option from the Period drop-down list.

Step 2 Choose an option from the Remove or Add column drop-down list.

Step 3 Click Submit. The changes are implemented.

Graphs

A table of graphs is displayed on the web pages where each row represents a gateway or a unit and each column matches one of the graph types described in Graph Types. Clicking a graph for a gateway, unit, or AVM opens the individual graph of the selected item in a separate page.

Each graph displays the IP address and name of the graph. For more information about graph types, see Graph Types. In addition, a key is displayed at the bottom of the graph.

There are two colored vertical lines (indicators) that can be displayed in the graphs (see Figure 12-3):

•An out-of-memory indicator (a red vertical line) is displayed on the Java heap graph when an AVM runs out of memory. This is displayed in the combined and individual AVM graphs.

•A restart indicator (a green vertical line) is displayed in the AVM graph of the specific individual AVM that was restarted, as well as in the combined graph, when a Cisco ANA unit is restarted.

Figure 12-3 Out of Memory and Restart Indicators

The horizontal lines displayed in the graphs show the maximum and minimum values for a period of time.

The graph history is maintained for a period of 28 days. The data history is maintained as follows:

•Data of every 15 seconds is saved for a period of 3 hours.

•All data from 3 hours to 24 hours is diluted to a sampling rate of 300 seconds.

•All data from 24 hours to 7 days is diluted to a sampling rate of 15 minutes.

•All data from 7 days to 28 days is diluted to a sampling rate of 2 hours.

Graph Types

The following types of graphs are available:

•Java Heap—The sizes of the Java heaps in the AVMs processes.

•Process Size—AVM memory process sizes.

•CPU %—AVM CPU usage.

•GC Time—AVMs Java Garbage Collector (GC) activity.

•Dropped Messages—The number of messages dropped in the Cisco ANA transport messaging mechanism. This can happen when the system is under a heavy load.

•Logged Lines—The number of lines written to the AVM logs.

•CPU Total—The system CPU metrics for Cisco ANA unit operation.

Web Page Options

You can view the following web pages:

•MC Loads Page—Displays a combined graph of the loads for the gateways and units.

•MC server-ip Load Page—Displays combined and individual AVM graphs.

We recommend that you do not use the following options:

•Transport Counters Page

•Status

The web page options that are available depend on the web page being viewed.

MC Loads Page

Each row in the table corresponds to a Cisco ANA unit or gateway, and each column matches one of the graph types described in Graph Types. The MC Loads page displays the combined unit and gateway graphs, where each graph stacks the readings from all AVMs in the unit graphs.

Clicking the hyperlink of a unit or gateway opens the MC [server ip] Load page. For more information, see MC server-ip Load Page.

MC server-ip Load Page

The first row in the table of the MC server-ip Load page displays the combined AVM graph, and every row thereafter represents the graphs for an individual AVM. Each graph stacks the readings for the AVM. See Figure 12-4.

Figure 12-4 MC [server ip] Load Page

Invoking Additional Parameters

When a single graph page is opened, additional parameters can be invoked on the graph through the browser URL field, in an HTTP GET format:

•period—The period of the graph, in hours, minutes, days, and weeks, by using the letters h, m, d, and w to indicate hours, minutes, days, and weeks respectively.

For example, to generate a graph that covers a period of three hours:

https://10.56.56.26:1311/graphs/graph.cgi?type=heap&mcip=127.0.0.1&period=3h

•end—The starting time of the graph, using the same time format as in "period."

For example, to generate a graph from two days ago for a period of four hours:

https://10.56.56.26:1311/graphs/graph.cgi?type=heap&mcip=127.0.0.1&period=4h&end=-2d

•refresh—Refreshes the graph page periodically. The period is defined in seconds. We recommend that you set the minimum period to 20 seconds, because Cisco ANA graph data is collected every 20 seconds.

In the following example, the page is refreshed every 20 seconds:

https://10.56.56.94:1311/graphs/graph.cgi?type=heap&mcip=127.0.0.1&refresh=20

•width—The width of the graph in pixels.

•height—The height of the graph in pixels.

In the following example, the graph is drawn with a width of 800 pixels and height of 600 pixels:

https://10.56.56.94:1311/graphs/graph.cgi?type=heap&mcip=127.0.0.1&width=800&height=600

Use Cases

This section provides two use case examples for the system health and diagnostics tool.

Use Case 1

If a unit CPU graph shows high CPU consumption and the GC Time graph is high as well, one of the AVMs might not have sufficient memory; for example, if the CPU is loaded with Java Garbage Collector tasks. If this is the case, reducing the Java Garbage Collector can help return CPU consumption to normal.

Use Case 2

For memory consumption, we recommend that 30% of the AVM memory remains free (in a steady state). The graphs provide you with a visual way to check this rate.

Check the Java Heap graph for the unit. The upper horizontal line displays the configured memory size for the AVM. The graph shows the real memory usage of the selected AVM.

For example, in the Java Heap graph shown in Figure 12-5, there is only 20% free memory:

•Configured memory: 250 Mb

•Used memory: 200 Mb

•Free Memory: 50 Mb (250-200)

Figure 12-5 Use Case 2