Contents

1   Overview

This document describes the JavaCAF PM Report Failure alarm and the procedure to follow when it is displayed in the alarm list.

1.1   Description

The JavaCAF PM Report Failure alarm is issued by the Performance Management (PM) sub-component when report files cannot be created in the output directory. The alarm attributes are listed in Table 1.

Table 1    Alarm Attributes

Attribute Name	Attribute Value
Major Type	com.ericsson
Minor Type	caf.pm.reportfailure
Managed Object Class	JavaCafPm
Specific Problem	JavaCAF, PM Report Failure
Event Type	org.saforum.ntf.alarm.communication (map to JavaCAF: SafAlarmType.COMMUNICATION)
Perceived Severity	Major (map to JavaCAF: SafAlarmSeverity.MAJOR)

The alarm contains information related to the following Attribute-Value Pair (AVP):

Table 2   

Attribute	Value	Description
outputPath	String	Defines the path where the PM report files are stored.

The AVP can be modified in the Information Model Management (IMM). For more information on the attributes, refer to the JavaOaM Managed Object ModelJavaOaM Managed Object Model, Reference [1].

The following are possible causes for the alarm condition:

• Missing output directory
• Insufficient access rights to the output directory
• Insufficient disk space where the output directory is located
• Inability to set the measurement report file group-owner to the javaoam-pm-data group when the javaoam-pm-data group must be used

Note:  

As a result of the fault, the report file cannot be written.

1.2   Prerequisite

For the definition of acronyms and terms, refer to the Glossary of Terms and AcronymsGlossary of Terms and Acronyms, Reference [2].

2   Procedure

To clear the alarm:

1. Print the file path to the report files, outputPath, from IMM.
2. Ensure that the file path exists.

   cd <parent directory outputPath>

3. Verify that the correct access rights are assigned.
   1. ls -al
   2. Verify that the printout includes the directory.
   3. Ensure that the directory has at least the access rights drw-------.
4. Ensure that there is sufficient disk space remaining for the report output:

   df /cluster

5. Ensure that the application process user is a member of javaoam-pm-data, if the javaoam-pm-data group exists on the system.
   1. Verify if the javaoam-pm-data group exists on the system.

      cat /etc/group | grep javaoam-pm-data

      Note:  

      The javaoam-pm-data group exists on the system if the printout includes javaoam-pm-data.

      
      
   2. If the javaoam-pm-data group exists on the system, verify if the application process user is a member of the javaoam-pm-data group.

      id <application process user>

      Note:  

      The application process user is a member of the javaoam-pm-data group if the printout includes javaoam-pm-data.

      
      
   3. If the javaoam-pm-data group exists on the system and the application process user is not a member of the javaoam-pm-data group, run the following command and then restart the application.

      usermod -a -G javaoam-pm-data <application process user>

Note:  

To help identify the root cause of the issue, the application logs can be verified for error messages. If the alarm does not clear or is issued frequently, consult the next level of maintenance support. Further actions are outside the scope of this guide.

Reference List

[1] JavaOaM Managed Object Model, MANAGED OBJECT MODEL, 155 54-LZN 708 0464/2-V1
[2] Glossary of Terms and Acronyms, TERMINOLOGY, 0033-APR 901 0487/2