Managing your Orchestrator logs

Orchestrator activity logs are useful for internal troubleshooting, as well as for sharing with IBM Aspera for technical support. You can set a policy for how Orchestrator retains log files.

As of Orchestrator 4.1.0, logs are written to both stdout and the local file production.log. In systemd-based environments such as RHEL 9.4, this means you can view logs in the following ways:
  • From the directory /opt/aspera/orchestrator/log/production.log. This directory also contains logs from previous days.
  • In the Orchestrator UI, by clicking Engine, then clicking Log Viewer in the left navigation.
  • Using journalctl to view logs per service.
Orchestrator retains logs based on a maximum total storage cap of 500 MB, as defined in the production.rb configuration file. When this limit is reached, the oldest logs are deleted automatically to make room for new entries.

The file orchestrator.log always contains the most recent log data.

For an explanation of how Orchestrator retains activity logs, and how to set a retention policy, see "Log Rotation in Orchestrator".

Log rotation in Orchestrator

To prevent your log file from growing too large, Orchestrator uses a total log storage cap of 500 MB. Logs are rotated once this limit is reached, and older logs are removed automatically.

The file production.log always contains the most recent log data.

Orchestrator controls the total number of log files kept in the system at any given time by deleting them after they have reached a maximum number of files.

Retrieving the Orchestrator logs

There are multiple ways to retrieve the Orchestrator logs.

Using journalctl (manual method)

You can run the following command to view the Orchestrator server logs directly through journalctl:

journalctl -u orchestrator_PROCESS
Using the orchestrator logs command (preferred method)

The orchestrator logs command provides a more convenient way to access Orchestrator logs.

  • To obtain logs for all services, run:

    orchestrator logs

  • To obtain logs for a specific service, run:

    orchestrator logs PROCESS

    Possible values to replace PROCESS in the commands include:

    • server
    • process_manager
    • engine
    • worker_1, worker_2

Viewing Orchestrator logs (journalctl)

Note: The -f flag follows the log output in real time, -n100 displays the last 100 log entries, and --no-pager disables the pager so the logs are printed directly to the terminal.

You can use journalctl commands on the host system to view logs for each Orchestrator service.

Server logs
journalctl -u orchestrator_server --no-pager -f -n100
Engine logs
journalctl -u orchestrator_engine --no-pager -f -n100
Monitor logs
journalctl -u orchestrator_monitor --no-pager -f -n100
All worker logs
journalctl -u orchestrator_worker_* --no-pager -f -n100
Logs for an individual worker

Replace <worker ID> with the ID shown in the Engine Processes page.

journalctl -u orchestrator_worker_<worker ID> --no-pager -f -n100