Logs

OMi records the procedures and actions performed by the various components in log files. The log files are usually designed to aid HPE Software Support when OMi does not perform as expected.

You can view log files with any text editor.

Most log files are located in the <OMi_HOME>/log directory and in subdirectories organized by component.

Log file properties are defined in files in the following directory and its subdirectories: <OMi_HOME>/conf/core/Tools/log4j.

In single-server installations, all OMi servers and their logs reside on the same machine. In the case of a distributed deployment of the servers among several machines, logs for a particular server are usually saved on the computer on which the server is installed. However, if it is necessary for you to inspect logs, you should do so on all machines.

When comparing logs on client machines with those on the OMi server machines, keep in mind that the date and time recorded in a log are recorded from the machine on which the log was produced. It follows that if there is a time difference between the server and client machines, the same event is recorded by each machine with a different time stamp.

Each log is configured so that the information it records corresponds to a certain severity threshold. Because the various logs are used to keep track of different information, each is preset to an appropriate default level. For details on changing the log level, see How to Change Log Levels.

Typical log levels are listed below from narrowest to widest scope:

• Error. The log records only events that adversely affect the immediate functioning of OMi. When a malfunction occurs, you can check if Error messages were logged and inspect their content to trace the source of the failure.

• Warning. The log's scope includes, in addition to Error-level events, problems for which OMi is currently able to compensate and incidents that should be noted to prevent possible future malfunctions.

• Info. The log records all activity. Most of the information is routine and the log file quickly fills up.

• Debug. This level is used by HPE Software Support when troubleshooting problems.

The default severity threshold level for log files differs per log, but is generally set to either Warning or Error.

A size limit is set for each type of log file. When a file reaches this limit, it is renamed and becomes an archived log. A new active log file is then created.

For many logs, you can configure the number of archived log files that are saved. When a file reaches its size limit, it is renamed with the numbered extension 1 (log.1). If there is currently an archived log with the extension 1 (log.1), it is renamed to log.2, log.2 becomes log.3, and so on, until the oldest archived log file (with the number corresponding to the maximum number of files to be saved) is permanently deleted.

The maximum file size and the number of archived log files are defined in the log properties files located in <OMi_HOME>/conf/core/Tools/log4j. An example is:

def.file.max.size=2000KB
def.files.backup.count=10

The log file <OMi_HOME>/log/jboss7_boot.log logs start-up activities including running the application server process, deployment, and start-up status, as well as the number of busy ports.

*.hprof files contain a dump heap of an OMi process's data structures. These files are generated by the Java virtual machine (JVM) if a process fails with a Java Out Of Heap Memory condition.

You are rarely aware of a problem because the problematic process restarts automatically after a failure. The existence of many *.hprof files indicates that there may be a problem in one of the OMi components, and its contents should be analyzed to determine the problem.

If you run out of disk space, you can delete the*.hprof files.

You can enable debug trace logging for an event by setting the custom attribute __TRACE__. It may have any severity level. By default, the event flow trace logging level is set to INFO. Only events with the custom attribute __TRACE__ set are logged to the flow trace log files. To enable flow tracing for all events, set the flow trace log level to DEBUG.

You can enable trace logging on the OM server or agent sending the event, or you can add the trace to the event at a later time. Whenever this custom attribute is enabled on an event, trace output for this event appears in the following trace logs:

• OMi data processing server: log/opr-backend/opr-flowtrace-backend.log

• OMi gateway server: log/wde/opr-gateway-flowtrace.log

If you want HPE support to look at the trace, click the export to log4j button to save it.

The Logging Administrator tool enables you to temporarily modify the level of details displayed in OMi logs, as well as create custom logs. You can access the OMi Logging Administrator Tool from the following URL:

http://<OMi Gateway FQDN>/topaz/logAdminBsm.jsp

You can delete all OMi log files under /opt/HB/BSM/log and *.hprof files under /opt/HP/OMi/bin after stopping OMi. This enables you to free up disk space. However, from a support perspective, it may be useful to save older logs.

1. Stop OMi.
2. Delete all files under <OMi_HOME>/log. Do not delete the log directory.
3. Delete all .hprof files under /opt/HP/BSM/bin/.

If requested by HPE Software Support, you may have to change the severity threshold level in a log, for example, to a debug level.

1. Open the log properties file in a text editor. Log file properties are defined in files in the following directory: <OMi_HOME>/conf/core/Tools/log4j.

2. Locate the loglevel parameter. For example,

   loglevel=ERROR

3. Change the level to the required level. For example,

   loglevel=DEBUG

4. Save the file.