Table of Contents

Troubleshooting

Troubleshooting

This section provides troubleshooting information for Resource Manager. This appendix includes the following sections:

• Checking Files After Installation
  
  
• Understanding Installation Error Messages
  
  
• Accessing the Resource Manager Server
  
  
• Adding Device Information
  
  
• Checking the Syslog Configuration
  
  
• Using Software Management
  
  
• Gathering Server Information
  
  
• Collecting Information Before Requesting Support
  
  

Checking Files After Installation

If you encountered problems during the installation of Resource Manager, make sure the following files have been installed. The following directories should be in the /opt/CSCOpx/ directory if the installation was successful:

• /bin
  
  
• /cgi-bin
  
  
• /conf
  
  
• /etc
  
  
• /example
  
  
• /htdocs
  
  
• /lib
  
• /man
  
  
• /objects
  
  

In addition, the following files should have been added to the /etc directory:

• rc2.d/K90dmgtd
  
  
• init.d/dmgtd
  
  
• rc.config.d/ciscoRMCtrl
  
  

The Resource Manager installation tool installed the following directories in the /opt/CSCOpx/objects directory:

• availability (Availability)
  
  
• inventory (Inventory)
  
  
• sysloga (Syslog Analysis)
  
  
• web (web server process and utilities)
  
  
• swim (Software Management)
  
  
• db (Resource Manager datastore)
  
  
• dmgt (daemon management)
  
  
• tcltk (web administration)
  
  
• util
  
  
• perl (Resource Manager perl interpreter and libraries)
  
  
• share
  
  
• data
  
  

Understanding Installation Error Messages

After making sure the correct files were installed, check the /opt/CSCOpx/ciscoinstall.log file for installation errors. You might find the following types of messages:

• Warning messages, which indicate that something might be wrong with a particular process, but the process will complete.
  
  
• Error messages, which indicate that a particular process could not complete.
  
  
• Info messages, which give you details that you need to know.
  
  

The following table lists error messages that might occur during installation and describes the reasons for the errors:

<sub-package> did not install.

Missing <sub-package> should 
exist in :<directory> 

Error: base package: <name> 
did not install. Exiting. 

pkgchk <pkg_name> failed.

<dir_name> not found. No 
additional packages 
installed.

cd <dir_name> failed. 
Applications not installed.

WARNING: RAM in system is 
$RAM. $MIN_RAM recommended.

WARNING: SWAP in system is 
less than 2x RAM.

ERROR: You must be root to run 
Solaris install. Exiting.

ERROR: Missing directory 
CSCOpx in $PWD.

Accessing the Resource Manager Server

Resource Manager automatically installs a web server during the initial installation. This web server uses the default port 80, which is normally used by web servers. If you receive an error message that an existing web server is already configured to on port 80, and the alternate port 1741 is used instead, make sure you specify the correct machine name:

server_name:1741 
 

(where server_name is the name of the machine where Resource Manager was installed, and 1741 is the alternate port on which Resource Manager is installed if port 80 was already in use). Use the following command to make sure your server is running:

ping server_name
 

If you get a message that the server is "alive," and get a proxy error when you try to connect to the server, make sure the proxy is set up correctly. If your server is configured to use a proxy server outside the firewall (specified in Netscape Navigator, under Options > Network Preferences > Proxies), you will get proxy errors if you have incorrectly set up the proxy to ignore requests to a certain machine, set of machines, or domain.

You know that your proxy is set up incorrectly if you encounter any of the following errors:

• You receive an error message that you are using a proxy outside the firewall.
  
  
• The proxy server recognizes www-int as an internal server, so it does not proxy requests to that server.
  
  
• You set up a new internal server, www-nms, but when you make a request to the proxy server, it doesn't recognize www-nms as an internal server and proxies the request.
  
  
• The proxy server outside the firewall tries to request data from a server inside the firewall, and the request is blocked.
  
  
• You get a "Connection Refused" error from the proxy server.
  
  

Setting Up the Browser

If the Resource Manager buttons do not work, you have not enabled Java and JavaScript. Follow these steps to enable Java and JavaScript:

• For Netscape Navigator 3.01, make sure Enable Java and Enable JavaScript are checked under Options > Network Preference > Languages.
  
  
• For Microsoft Internet Explorer 3.01 and 3.02, make sure Enable Java Programs and Active-X Scripting are enabled under View > Options > Security.
  

Make sure your cache is not set to zero. If you experience browser problems, increase your disk cache setting. Do not run multiple copies of Netscape Navigator on Solaris.

Do not resize the browser window while the desktop or main page is still loading. This will cause a Java error.

Logging In

If you are logging in as an administrator and you receive the following error message from the browser:

Authorization Required. This server could not verify that you are authorized to access the document . . .
 

you have not entered a valid user name or password, or you have clicked Cancel on the User Name and Password Required dialog box. Make sure you specify a valid administrator user name and password.

Adding Device Information

If you added a device using Inventory > Add Devices and the Add/Import Status Summary screen indicates that the device status has not changed from pending within 15 minutes, make sure you check the status of all processes to ensure that they are running normally, as explained in the following steps:

Step 1 To view the latest device status information, in the Add/Import Status Summary screen (Inventory > Import Status), click Update.

Step 2 To check if the DIServer is running, open the System Admin folder, and click Process Status. (The DIServer is the process responsible for validating devices and changing their status from pending.)

If the DIServer process has the state running normally, it might be caught in an infinite loop. You need to stop and restart it by following these steps:

To restart the DIServer, follow these steps:

Step 3 Return to the Add/Import Status Summary screen by clicking Inventory > Import Status and click Update. The device status should change to managed within a couple minutes.

Checking the Syslog Configuration

If the Resource Manager installation fails while attempting to install Syslog Analysis, make sure you do not have the Cisco DNS/DHCP Manager (CDDM) installed on the same machine as Resource Manager. CDDM, which can be downloaded from CCO, provides syslog and TFTP daemons as part of its product. You cannot run CDDM Syslog or TFTP on the same machine as Resource Manager because they are not compatible. Make sure that the server on which you install Resource Manager is not running CDDM Syslog or TFTP.

Using Software Management

If you install Software Management for the first time, and the /tftpboot directory is a remote directory, the following message shows during the installation of Software Management:

## Executing postinstall script.
INFORMATIONAL: The tftp directory '/auto/tftpboot' appears to be NFS mounted. 
NFS directories often result in slower software transfers, which might cause the software transfer functionality in Software Management to fail Ask the System Administrator to make /auto/tftpboot a local directory.
 

You should make /auto/tftpboot a local directory.

If you have not removed the user bin from the /usr/lib/cron/at.deny file, the following warning message appears during the installation:

WARNING: User 'bin' appears in '/usr/lib/cron/at.deny'. The software transfer jobs require 'bin' to be able to run 'at'. Ask the System Administrator to remove bin from /usr/lib/cron/at.deny.
 

You need to remove bin from the /usr/lib/cron/at.deny file.

Gathering Server Information

Resource Manager contains a utility that can help you troubleshoot problems with the server. To get information about the Resource Manager server, follow these steps:

Step 1 Click Admin > System Admin > Collect Server Info. The Collect Server Information dialog box appears.

Step 2 Click Finish.

Information about the Resource Manager sever is collected. After the information has been collected, you receive a message indicating where the server status information file is located (/var/tmp/px_status_info).

You need this file when requesting support from Cisco, as explained in the next section, "Collecting Information Before Requesting Support."

Collecting Information Before Requesting Support

To help determine the cause of any problems you encounter, your technical support representative might request that you gather product information. To help you gather this information, Resource Manager contains a tool that gathers server information, which can be accessed in one of two ways:

• Click Admin > System Admin > Collect Server Info. The Collect Server Information dialog box appears. Click Finish, and the server information will be collected into the /var/tmp/px_status.info file.
  
  
• Enter the following command:
  

# /opt/CSCOpx/bin/collect.info
 

A px_status.info file is created in the /var/tmp directory. If the file is not created, make sure you are logged in as root and that you source the /opt/CSCOpx/etc/install.cshrc or the /opt/CSCOpx/etc/install.profile file.

E-mail this file to Cisco if directed to do so by your technical support representative.