MinimServer Forum

Full Version: high CPU usage by tail after initiating rescan
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Hi,

I have a recently changed behaviour with MinimServer on my FreeBSD-installation. I think it's just a minor issue, but I fail in finding out what's causing it.

In my minimserver-installation, I have a process in my process-list, which reads like

Code:
root         11169   0.0  0.0    9952   1544 v0- IJ    9:47PM  0:00.00 tail -n 1 -f /usr/local/share/minimserver/data/minimserver.pipe

This process is always present and doesn't consume a lot of CPU (here: 0).

It's only after I initiate a rescan, when this process becomes a problem. I initate rescan always automatically when files are added by executing the following command as a non-root user.

Code:
airflow@services_2:/ % echo "rescan" > /usr/local/share/minimserver/data/minimserver.pipe

Rescan happens and works normally. MinimServer works fine with the added files.

It's just that the tail-process consumes 100% CPU forever after that. When I shut the complete system down and start MinimServer normally with the start-scripts, everything is fine again - until I trigger the next rescan with above method. Is there a bug, or do I have to change something with my method? This has not been like this before, but there were updates with FreeBSD and MinimServer, and I'm not sure what's causing it.

Code:
root    11169 100.0  0.0    9952   1560 v0- RJ    9:47PM 8:16.73 tail -n 1 -f /usr/local/share/minimserver/data/minimserver.pipe

Thanks,
airflow
Thanks for letting me know about this.

I have tried this on Linux and I am not seeing this problem.

There will be a new version of MinimServer soon that will provide a new way of doing scripted rescan commands without the need to use a named pipe that is monitored by a tail process. I hope this will resolve the issue.
(13-04-2016 23:04)simoncn Wrote: [ -> ]There will be a new version of MinimServer soon that will provide a new way of doing scripted rescan commands without the need to use a named pipe that is monitored by a tail process. I hope this will resolve the issue.

I just tested the new command scripting facility and it works fine. This also solved the problem with high CPU described in this thread. Thanks!
Reference URL's