CPU usage 0.0% just after process start.

[mschriber]

If a process has a minimum CPU usage defined ("if cpu < 5% then alert"), the check will fail just after the process has been started, because the cpu usage seems to be always 0.0% for the first check.

Having the option to define a minimum CPU usage, would allow to monitor certain processes in greater detail.

[EDT Oct  8 17:46:01] info     : 'cpu_test' start on user request
[EDT Oct  8 17:46:01] info     : Monit daemon with PID 10411 awakened
[EDT Oct  8 17:46:01] info     : Awakened by User defined signal 1
[EDT Oct  8 17:46:01] info     : 'cpu_test' start: '...'
[EDT Oct  8 17:46:01] info     : 'cpu_test' start action done
[EDT Oct  8 17:46:01] error    : 'cpu_test' cpu usage of 0.0% matches resource limit [cpu usage<5.0%]

[Tildeslash]

Please can you run monit in debug mode and send output?:

monit -vI

The cpu usage test should be skipped while it's initializing.

[mschriber]

Output of the test:

Adding 'allow localhost' -- host resolved to [::ffff:127.0.0.1]
Runtime constants:
 Control file       = /home/mschriber/monit/cpu_test/monitrc
 Log file           = /home/mschriber/monit/cpu_test/monit.log
 Pid file           = /home/mschriber/.monit.pid
 Id file            = /home/mschriber/.monit.id
 State file         = /home/mschriber/.monit.state
 Debug              = True
 Log                = True
 Use syslog         = False
 Is Daemon          = True
 Use process engine = True
 Limits             = {
                    =   programOutput:     512 B
                    =   sendExpectBuffer:  256 B
                    =   fileContentBuffer: 512 B
                    =   httpContentBuffer: 1024 kB
                    =   networkTimeout:    5 s
                    =   programTimeout:    5 m
                    =   stopTimeout:       30 s
                    =   startTimeout:      30 s
                    =   restartTimeout:    30 s
                    = }
 On reboot          = start
 Poll time          = 3 seconds with start delay 0 seconds
 Start monit httpd  = True
 httpd bind address = localhost
 httpd portnumber   = 2813
 httpd ssl          = Disabled
 httpd signature    = Enabled
 httpd auth. style  = Host/Net allow list

The service list contains the following entries:

Process Name          = cpu_test
 Pid file             = /home/mschriber/monit/cpu_test/pid
 Monitoring mode      = active
 On reboot            = start
 Start program        = '/bin/bash -c /home/mschriber/monit/cpu_test/run.py &' timeout 30 s
 Stop program         = '/bin/bash -c kill `cat /home/mschriber/monit/cpu_test/pid`; sleep 1' timeout 30 s
 Existence            = if does not exist then restart
 CPU usage limit      = if less than 5.0% then alert

System Name           = h01
 Monitoring mode      = active
 On reboot            = start

-------------------------------------------------------------------------------
pidfile '/home/mschriber/.monit.pid' does not exist
Starting Monit 5.20.0 daemon with http interface at [localhost]:2813
Starting Monit HTTP server at [localhost]:2813
Monit HTTP server started
'h01' Monit 5.20.0 started

'cpu_test' start on user request
Monit daemon with PID 24795 awakened
Awakened by User defined signal 1
pidfile '/home/mschriber/monit/cpu_test/pid' does not exist
'cpu_test' start: '/bin/bash -c /home/mschriber/monit/cpu_test/run.py &'
'cpu_test' started
'cpu_test' monitoring enabled
'cpu_test' start action done
'cpu_test' process is running with pid 24801
'cpu_test' zombie check succeeded
'cpu_test' cpu usage of 0.0% matches resource limit [cpu usage<5.0%]
'cpu_test' process is running with pid 24801
'cpu_test' zombie check succeeded
'cpu_test' cpu usage check succeeded [current cpu usage=99.9%]
'cpu_test' process is running with pid 24801
'cpu_test' zombie check succeeded
'cpu_test' cpu usage check succeeded [current cpu usage=99.9%]
'cpu_test' process is running with pid 24801
'cpu_test' zombie check succeeded
'cpu_test' cpu usage check succeeded [current cpu usage=100.0%]

'cpu_test' stop on user request
Monit daemon with PID 24795 awakened
Awakened by User defined signal 1
'cpu_test' stop: '/bin/bash -c kill `cat /home/mschriber/monit/cpu_test/pid`; sleep 1'
'cpu_test' stopped
'cpu_test' monitoring disabled
'cpu_test' stop action done

The script goes into an infinite loop right away after the PID-file was created.

Log file part with time-stamps and log severity:

[EDT Oct 10 04:12:36] info     : 'cpu_test' start on user request
[EDT Oct 10 04:12:36] info     : Monit daemon with PID 24795 awakened
[EDT Oct 10 04:12:36] info     : Awakened by User defined signal 1
[EDT Oct 10 04:12:36] debug    : pidfile '/home/mschriber/monit/cpu_test/pid' does not exist
[EDT Oct 10 04:12:36] info     : 'cpu_test' start: '/bin/bash -c /home/mschriber/monit/cpu_test/run.py &'
[EDT Oct 10 04:12:36] debug    : 'cpu_test' started
[EDT Oct 10 04:12:36] debug    : 'cpu_test' monitoring enabled
[EDT Oct 10 04:12:36] info     : 'cpu_test' start action done
[EDT Oct 10 04:12:36] debug    : 'cpu_test' process is running with pid 24801
[EDT Oct 10 04:12:36] debug    : 'cpu_test' zombie check succeeded
[EDT Oct 10 04:12:36] error    : 'cpu_test' cpu usage of 0.0% matches resource limit [cpu usage<5.0%]
[EDT Oct 10 04:12:39] debug    : 'cpu_test' process is running with pid 24801

Many thanks.

[Tildeslash]

Fixed: Issue #676: Process CPU usage is reported always as 0.0% just after process start

→ <<cset 248c17414bba>>

[Tildeslash]

Thanks for data, the problem is fixed.

If you want to test the development version snapshot:

wget https://bitbucket.org/tildeslash/monit/get/master.tar.gz
tar -xzf master.tar.gz
cd tildeslash*
./bootstrap
./configure
make

[mschriber]

Many thanks for fixing it so quickly!