WAS v8.5 > Reference > Log and trace file descriptionsHPEL trace configuration settings page
We can only access this page when the server is configured to use HPEL log and trace mode.
To view this dmgr console page, click...
Troubleshooting | Logs and trace | server_name| Configure HPEL trace
Trace to a directory
Tracing system writes trace records to the trace directory as they are created by the server.
Enable log record buffering
Tracing system avoids writing to disk each time a trace record is created. The tracing system creates a buffer that can hold a number of trace records, and writes the buffered events when the buffer is full. The tracing system also writes the buffered events after a few seconds have passed, even if the buffer is not full.
Selecting this setting significantly improves tracing performance; however, if the server stops unexpectedly, the contents might not be written to the trace repository.
Best practice: Enable trace record buffering in almost all cases. Only disable trace record buffering when the server is failing unexpectedly and cannot write buffered content to disk before stopping..
Start new log file daily at <time>
Enables the tracing framework to close the trace file and start a new file at the specified time of day. Closing the file makes it easy to copy the file to an archive.
Best practice: To automatically archive your trace files, set up your backup program to copy files after the time you configured for new files to be started. Configure the backup to occur at least 10 minutes after the time configured for new files to be started to ensure the server has closed the previous file.
Begin cleanup of oldest records
Trace cleanup settings to be used to automatically purge the oldest trace records, or trace records that no longer fit in the configured space, from the trace repository.
Select When log size approaches maximum to configure automatic trace file cleanup to begin when the total size of the trace repository approaches the configured maximum size.
Select When oldest records reach age limit to configure automatic trace file cleanup to begin when trace content is the age limit specified.
Select When either age or size restriction is met to configure automatic trace file cleanup to begin when either of the previous conditions is met.
Regardless of the selection chosen, records are deleted from the trace repository in the order in which they were written to the trace repository.
Log record age limit
Lifespan, in hours, that trace records can remain in the trace repository before the trace records can be automatically deleted by the server. When the oldest records in the trace repository have existed longer than the age limit specified, then those records are targeted for deletion by the server.
Maximum log size
Maximum total size, in megabytes, the server allows the trace repository to reach. When the trace repository approaches this size limit, the server deletes the oldest records from the trace repository to make space for new trace records.
Out of space action
How the server reacts to an inability to add content to the trace repository.
Select Stop server to specify the server stops when the server is unable to write to the trace repository.
Select Purge old records to specify the server continues to run, and the oldest trace records are immediately removed when the server is unable to write to the trace repository.
Select Stop logging to specify the server continues to run, but the server cannot continue to write to the trace when the server is unable to write to the trace repository.
Trace to a memory buffer
Tracing system writes trace records to a memory buffer.
We can write the memory buffer contents to the trace directory from the runtime tab.
Memory buffer size
Amount of memory the tracing system allocates in the server to contain trace records.
In cases where the memory buffer is full when a new trace record is created, the oldest entry from the memory buffer is deleted to make space.
Dump button
Use this button to write the contents of the trace memory buffer to the trace directory.
The tracing system resets the memory buffer after you dump it. The tracing system continues to record trace records in the memory buffer after you dump the buffer.
Directory to use for tracing and dumping memory buffer
Directory to which trace files are written. A subdirectory, trace data, is created in this directory, and the trace files are written to this location.
This path cannot be shared between servers.
The server creates a file with a .owner extension to help detect when two or more servers happen to be trying to use the same path for HPEL output.
Save runtime changes to configuration as well
Changes are made to both the dynamic state of the running server, and the server configuration, which takes effect on the next restart. If this check box is not selected, the server does not copy the settings into the server configuration.
Related
Change to HPEL logging and tracing from basic mode