We have headless Linux boxes on our network that we have KACE 6.4 agents running on that report to a K1000. We have started doing Managed Installs on them, having 6 pieces of software that install when a new Linux device comes online.

Each Managed Install works fine on their own. However, when all of them are queued, it seems to get stuck after every 2 installs. The packages it gets stuck on actually install and start up just fine, but the automatic runkbot 6 x process that spawns them seems to hang in the background, and the only ways I've found to make it proceed with the rest of the MIs is to either kill the runkbot process and then do a '2 0', or reboot the device. I have the MIs ordered, and I've tried changing the order they get installed to pinpoint exactly which package it may be getting stuck on, but there doesn't seem to be a pattern.

These installations are not standard rpm installs using yum; they use a custom command line specified in their MI entry in the KACE web interface, most involving several commands stacked and separated by either '&&' or ';' with some of the commands calling on a shell script that is bundled in each respective MIs Software Inventory entry. I made sure that the shell scripts I'm using end with an 'exit' command and even threw in "2>&1" at the end of the commands in case there's a stderr stream keeping the process tied up in the background.

Has anyone seen this behavior before? Is there anything else I can look for in my scripts or install commands that could be causing this?
0 Comments   [ - ] Hide Comments


Please log in to comment

There are no answers at this time
Answer this question or Comment on this question for clarity