If you’re running a multi-context, on-premises Jamf Pro environment, you probably have some kind of third-party log aggregation and reporting tool to hand. But if you don’t have that yet, here’s a simple script you can use to get an aggregated output of errors and severe warnings from all your Jamf instances to a Slack channel.

It will report all log output marked as [ERROR] or [SEVERE] in the previous hour.


This script assumes you are running Jamf Pro on a Linux server and have anacron installed. It also assumes that tomcat instance logs are reported to /var/log/JSS/<instance>/JAMFSoftwareServer.log However, it would work from any task scheduling tool such as crontab.

Simply provide a Slack incoming web hook for whichever channel you want to use, and place this script into /etc/cron.hourly. The script needs to be executable.

The script outputs errors with the time stamp. If the error is SEVERE, this is explicitly printed in the output.

The script

Example output