Monit uses 65% of of CPU

[tht]

I noticed monit using up 65% of my CPU on my HP Mediasmart server. I checked /var/log/monit.log and found the only entries were for:

[EST Dec 23 13:02:58] error : 'vncserver-4' process is not running
[EST Dec 23 13:02:58] info : 'vncserver-4' trying to restart
[EST Dec 23 13:02:58] info : 'vncserver-4' start: /usr/bin/systemctl
[EST Dec 23 13:03:28] error : 'vncserver-4' failed to start
[EST Dec 23 13:03:28] error : 'afpd' process is not running
[EST Dec 23 13:03:28] info : 'afpd' trying to restart
[EST Dec 23 13:03:28] info : 'afpd' start: /usr/bin/systemctl
[EST Dec 23 13:03:58] error : 'afpd' failed to start

These entries happen every minute and impact overall system performance. Any suggestions?
Amahi 7.2 on Fedora 19

[bigfoot65]

Do you have these apps installed on your HDA? Are the services currently running? You can remove the files from /etc/monit.d directory and restart monit to stop monitoring of these services.

If the apps are installed and running, there is something else wrong then.

[tht]

I uninstalled VNC. Amahi Time Machine is working correctly. It looks like monit is looking for a PID file for it that doesn't exist. I commented that check out of the monit.d/ config file as a workaround and it seems to have resolved the symptom. Any suggestions on fixing it?

[bigfoot65]

My recommendation would be to find out the pid file name and location. Then update the .conf file. That should allow things to work correctly.

[tht]

I figured this out and it resolved the cpu issues.

Edit /usr/libexec/netatalk/netatalk.sh
Change the following line:

Code: Select all

AFPD_UAMLIST= "-U uams_dhx.so,uams_dhx2.so”

to

Code: Select all

AFPD_UAMLIST="-P /var/run/afpd.pid -U uams_dhx.so,uams_dhx2.so”

Restart the netatalk.service
Restart the monit.service

[bigfoot65]

Would you mind adding this to the wiki? It would be good for users who experience the same problem. The Amahi wiki is our primary source of tutorials and how to's for Amahi vs the forums.