Troubleshooting


You can use the information in this topic to troubleshoot issues relating to data collection and Java folder configuration.

Data collection configuration

To make sure that BMC AMI Ops User Interface receives expected data, configure the following for BMC AMI Ops User Interface and the BMC AMI Ops products:

Java configuration

Your BMC AMI Ops User Interface job STDOUT log shows the following error message:

CEE3501S The module libjvm.so was not found.                                   
From entry point JzosVM::initializeVMArgs() at compile unit offset +000 00094 at entry offset +00000094 at address 4310620C. 

The error message indicates that the BMC AMI Ops User Interface environment is not correctly pointing to the Java 17 (64-bit) for one of the following reasons:

  1. You applied the latest PTF and the IBM SDK Java 17 (64-bit) is not in the default /usr/lpp/java/J17 directory.
  2. Your server is pointing to the correct Java directory, but updates to the Java package changed the Java SDK library location in your mainframe.

To resolve this error if you have not yet applied the latest PTF, apply the PTF and restart the server.

If you still see the error, your server is probably not in the default directory. Perform the following steps:

  1. Open the server started task and update the installation path for the BMC AMI Ops User Interface.
  2. Run the server started task and make sure it executes with return code 0.
  3. Restart the server started task for the BMC AMI Ops User Interface.

    • Default started task names by servers
      • AMIAPSRV—for application server
      • AMIDSSRV—for registry service
      • AMIMOSRV—for monitor service

Low Java Virtual Machine heap size

If you get the following error when working on a view:

AMI OPS UI / AMIMOSRV: "java/lang/OutOfMemoryError" at 2022/11/23 12:48:21 - please wait.

This occurs because the parameter for Java Virtual Machine (JVM) heap size is set to a min of 256m and a max of 512m as the default memory. When AMIMOSRV runs out of memory, the error message is displayed. runs out of memory the error message is displayed.

To increase the JVM heap size that is located in the BMCSAMP dataset in MUXMONEV member.

  1. In the following default parameter:

    IJO="$IJO -Xms256m -Xmx512m"
    • Replace -Xms256m with Xms512m
    • Replace -Xmx512m with -Xmx1048m
  2. Now stop the AMIMOSRV started task and restart.

    This restarts the server.

Accessing different components of BMC AMI Ops User Interface components in different environments in the same browser

This section describes how to move component connection from one BMC AMI Ops User Interface to other BMC AMI Ops User Interface, so that you can view this component in the same browser along with the existing components from the other BMC AMI Ops User Interface.

Connect both the components by using one of the following procedures.

Configuring the components from AMICMNEV member

To connect the component (BMC AMI Ops Monitor or BMC AMI ManagerBMC AMI Manager) from one environment (suppose Env#1) to another (suppose Env#2), configure the AMICMNEV member as follows: 

Important

Performing this procedure disables the BMC AMI Ops UI Server and BMC AMI Ops UI Discovery server server from Env #2 and these servers cannot be used anymore.

  1. Open AMICMNEV member in UBBSAMP in Env#2 for editing.
  2. Copy the following values from the UBBSAMP dataset's AMICMNEV member in Env#2 to the AMICMNEV member in Env#1:
    • JWT
    • AMIDSC_PORT
    • AMIDSC_PORT_TYPE
    • AMIDSC_HOST

      Important

      If the BMC AMI Ops UI Discovery server is on a remote system, use the BMC AMI Ops UI Discovery server server host name and not the localhost.

    • AMIDSC_SECURED
    • AMIDSC_UNSECURED

Configuring the components from JCL

Configure the component (BMC AMI Manager or BMC AMI Ops Monitor) from Env#2 JCL by pointing it to the common member in Env#1.

For example, change the AMICMNEV member in the BMC AMI Manager JCL in Env#2 UBBSAMP to the one that is used in AMICMNEV member in Env #1 UBBSAMP.

Important

Performing this procedure does not disable the BMC AMI Ops UI Server and BMC AMI Ops UI Discovery server server from Env #1

Insufficient authority issue

When setting up the started tasks required for the 

BMC AMI Ops User Interface

, an error occurs in the started task that disables the permission for the users to change the file attributes. To write to the log, a user must have the permission to Write. During installation, all the file systems have Read and Write permissions.

To change the permissions, perform the following steps:

Important

The User ID assigned to the started task (STC) must have an OMVS segment defined.

The assigned user must have the Read, Write, and Execute permission. 

  1. Enter tso omvs command on mainframe command prompt.
  2. Go to run time file system directory location.
    For example,</u/MAINVIEW/ospui>
  3. Enter chmod 775 <fileName>.
    For more information on chmod, see IBM's chmod - Change the mode of a file or directory topic.

Login issues with BMC AMI Ops Analytics Agent

If you are unable to login to BMC AMI Ops Analytics Agent by using the provided MVHost, and on the Product Health page, the following message is displayed:

Unable to login, if this issue persists, restart MVHost.

To resolve this issue, make sure that the following PTFs are applied to BMC AMI Ops Infrastructure:

Version

PTF

BMC AMI Ops Infrastructure 7.1

BQY3585

BMC AMI Ops Infrastructure 7.0

BQY3584

BMC AMI Ops Analytics Agent cannot connect to BMC AMI Ops Analytics Datastore

BMC AMI Ops Analytics Datastore

If the following error is displayed in the eureka-client.log file, restart BMC AMI Ops Analytics Agent:

java.util.concurrent.RejectedExecutionException: Task java.util.concurrent.FutureTask@1dacd8c1[Not completed, task = java.util.concurrent.Executors$RunnableAdapter@4b8f69eb[Wrapped task = com.netflix.discovery.DiscoveryClient$CacheRefreshThread@1ec80a6c]] rejected from java.util.concurrent.ThreadPoolExecutor@79451696[Running, pool size = 2, active threads = 2, queued tasks = 0, completed tasks = 0]

For information, see Starting-and-stopping-analytics-components.

Log4J2 input/output error

Log4J2 fails to write to log files after a dynamic aggregate ownership movement in z/OS and gives an EDC5122I Input/output error.

To resolve this issue, follow these steps:

  1. Add the following parameter under the <Configuration> tag in the log4j2.xml file:
    <Configuration monitorInterval="30">
    This enables Log4J2 to periodically detect and apply changes to the configuration file dynamically.

    Info

    <RollingFile name="RoutingRollingFile" fileName="logs/routing.log"
    immediateFlush="true" ... >
    <!-- Additional configurations for the appender -->
    </RollingFile>

  2. Add the following parameter to the relevant <RollingFile> appender in the log4j2.xml file:
    immediateFlush="true" Parameter
    This ensures that the log messages are written to the log file immediately, and there is no loss of data during the transition. Make sure that the appender used for adding the parameters are related to the affected log files.

    Info

    <RollingFile
    name={name}
    fileName="${LOGS_DIR}/nameOfFile.log"
    filePattern="${LOGS_DIR}/nameOfFile.%i.log"
    ignoreExceptions="false"
    immediateFlush="true">

    Warning

    When implementing this resolution, consider the following points:

    • Performance overhead: The monitorInterval attribute might increase CPU and I/O usage, especially with network-mounted log files.
    • Debugging challenges: Dynamic reloading can complicate troubleshooting because changes take effect immediately.
    • Unintended reloading: Configuration file corruption or accidental edits might cause unnecessary reloads.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*