FOGTaskScheduler daemon uses 100% of CPU

[Tom Elliott]

Can you check with:

systemctl -l status FOGScheduler

[Fernando Gietz]

# systemctl -l status FOGScheduler
● FOGScheduler.service - FOGScheduler
   Loaded: loaded (/usr/lib/systemd/system/FOGScheduler.service; enabled; vendor preset: disabled)
   Active: active (running) since lun 2017-06-19 18:44:45 CEST; 18h ago
 Main PID: 1136 (FOGTaskSchedule)
   CGroup: /system.slice/FOGScheduler.service
           ├─1136 /usr/bin/php -q /opt/fog/service/FOGTaskScheduler/FOGTaskScheduler &
           └─1140 /usr/bin/php -q /opt/fog/service/FOGTaskScheduler/FOGTaskScheduler &

jun 19 18:44:45 fog7.lgp.ehu.es systemd[1]: Started FOGScheduler.
jun 19 18:44:45 fog7.lgp.ehu.es systemd[1]: Starting FOGScheduler...

Doing a ps

# ps aux | grep FOGTas
root      1136  0.0  0.1 330196 17388 ?        Ss   jun19   0:00 /usr/bin/php -q /opt/fog/service/FOGTaskScheduler/FOGTaskScheduler &
root      1140 99.9  0.1 418668 15072 ?        R    jun19 1134:24 /usr/bin/php -q /opt/fog/service/FOGTaskScheduler/FOGTaskScheduler &

[Tom Elliott]

@Fernando-Gietz what’s your load average on the machine?

[Fernando Gietz]

[Wayne Workman]

@Fernando-Gietz Just a side question, what utility outputs all those columns like that?

For the load issue - I’d say check the sleep time? FOG Configuration -> FOG Settings -> FOG Linux service sleep times You could also view active queries in mysql by issuing show processlist; that has lead me to root causes with fog several times.

[Fernando Gietz]

Hi Wayne,

[Tom Elliott]

What’s the load average.

The Top you command you used is not showing me the actual load averages.

Maybe uptime would be better?

[Fernando Gietz]

I have started the FOGScheduler daemon and the load is:

[root@fog7 fogproject]# systemctl start FOGScheduler
[root@fog7 fogproject]# uptime
 16:54:22 up 6 days,  2:30,  1 user,  load average: 1,19, 0,78, 0,90
 16:54:28 up 6 days,  2:30,  1 user,  load average: 1,18, 0,78, 0,90
 16:54:35 up 6 days,  2:30,  1 user,  load average: 1,53, 0,87, 0,93
 16:54:42 up 6 days,  2:30,  1 user,  load average: 1,57, 0,89, 0,94
 16:55:39 up 6 days,  2:31,  1 user,  load average: 2,02, 1,13, 1,01
 17:11:10 up 6 days,  2:46,  1 user,  load average: 1,80, 1,63, 1,30

I stop the daemon and …

[root@fog7 fogproject]# systemctl stop FOGScheduler
[root@fog7 fogproject]# uptime
 17:14:05 up 6 days,  2:49,  1 user,  load average: 1,66, 1,65, 1,36
[root@fog7 fogproject]# uptime
 17:18:42 up 6 days,  2:54,  1 user,  load average: 0,86, 1,31, 1,31
[root@fog7 fogproject]# uptime
 17:20:26 up 6 days,  2:56,  1 user,  load average: 0,86, 1,16, 1,25

[Fernando Gietz]

Solved!! The problem was some tasks that had misconfigured the cron hour

[Tom Elliott]

@Fernando-Gietz How do you mean?

[Wayne Workman]

@Tom-Elliott like a cron task with all * I guess.

[Fernando Gietz]

Yes. In the hour field 0, 23/2