Tracer disk usage FAQ
For effective debugging, Tracer creates a file with the details of traced events so that you can make an informed decision about processes in your application. When you trace a high number of events, the file size can become extensive. To avoid issues with disk usage, Tracer sets a limit on the file size. As a result, tracing your application has no impact on running of your system.
To which deployments of Pega Platform does the limit apply?
The limit on disk usage applies to on-premises Pega Platform deployments, Pega Cloud deployments, and Pega Platform deployments on client-managed clouds.
What is the disk usage limit?
The disk usage limit is 10 GB per node. The limit is the same for all nodes in a system, and the disk limit check occurs at the node level. Multiple sessions on one node share the disk usage limit.
In a sample scenario, four Tracer sessions run on two nodes. Sessions S1 and S2 run on node N1, and sessions S3 and S4 run on node N2. The disk usage limit defines Tracer behavior in the following way:
- When the session S1 reaches the limit of 10 GB, S1 stops and the system deletes the respective file of event details.
- Sessions S2, S3, and S4 continue to run.
What happens when the file size reaches the disk usage limit?
When the event file reaches the size of 10 GB, Tracer behaves in the following way:
- Tracing of the session that reaches the disk usage limit stops.
- The system deletes the event file. You can then only download error event messages.
- In the Tracer window, you can view the header of the previous Tracer session without the full event details.
Does the system reclaim the disk space after stopping a Tracer session?
Yes, the system reclaims the disk space after the file size in a session reaches 10 GB.
Does the Tracer stop running after reaching the disk usage limit?
Yes, the Tracer stops a tracing session that reaches the disk usage limit. For service rule tracing, the system generates a disk write event. Consider a scenario in which sessions S1 and S2 run on the same node. The node already uses 9 GB of disk space. If session S1 tries to use another 1.5 GB, the system stops session S1. Session S2 continues to run.
How does service rule tracing work with the disk usage limit?
Service rule tracing continues to run even after the event files reach the size of 10 GB, however, the system deletes the file that exceeded the limit and you can view only event error messages.
Can I increase or decrease the disk usage limit for Tracer?
Yes, you can change the disk usage limit by creating and then editing the tracer/disk/maxusage dynamic system setting. The minimum value for the disk usage limit is 5 GB.
For more information about dynamic system settings, see Configuring dynamic system settings.
Does the system delete the event file when the file size reaches 10 GB?
Yes, the system deletes the event file after the file size reaches 10 GB. The Tracer file that you can later download contains only error messages.
Does each Tracer session have a separate event file?
Yes, each Tracer session has a separate event file. If you have multiple Tracer sessions that run on one node, the total maximum size of all event files is 10 GB.
How does the system maintain the disk usage limit?
The system maintains the disk usage limit at the node level. Multiple Tracer sessions share the usage when running on one node.
In a sample scenario, sessions S1 and S2 run on node N1, while session S3 runs on node N2. Session S1 occupies 9 GB in an event file, and session S2 uses 700 MB. Session S2 needs 200 MB to write details about the next event to trace. The system allows the action because the total space that sessions S1 and S2 need in the files is less than 10 GB. Next, session S2 tries to write details about the next event that require another 200 MB. The system stops session S2 and sends the message that the file size reaches the Tracer disk usage limit for node N1. The system deletes the event file for session S2 and users can then only view the error messages. In the meantime, session S3 tries to write event details that require one GB in the event file. The system allows the action because every node has a separate limit of 10 GB.
What happens if a Tracer session starts and a browser crashes, or the Tracer does not receive a close event before the event file size reaches 10 GB?
The Tracer session remains active and writes events to the file. However, the system treats the session as stale and deletes the session automatically after an hour. You can change the timeout value after which the system deletes stale sessions by editing the tracer/session/timeout dynamic system setting.
For more information about editing dynamic system settings, see Configuring dynamic system settings.
What can I do when the file size constantly reaches the 10 GB limit, even if I do not run any Tracer sessions that use disk space?
Verify with your system administrator that you close all Tracer sessions that are not necessary at the moment. The list of operators that run Tracer sessions is available in the error message after reaching the disk usage limit. Also, ensure that you select only necessary rules and events to trace instead of tracing all available events. If you do not need to view clipboard page content, selecting the Abbreviate Events check box in the Tracer settings drastically reduces the disk usage.
How can I reclaim disk space when another operator starts a Tracer session and forgets to close the session?
You can check the other operators who run Tracer sessions. The list of operators is available in the error message that you receive after the event file size reaches the disk usage limit. You can also close Tracer sessions from Admin Studio.
For more information, see Managing requestors.