process command parameters not working

i want to setup process service differently .

i don't want count of all process but only process which are using CPU more then 99 percentage .

i have configure server following way .

i can see defined Custom attributes in icinga web

still it count all running process .

how to configure process service so it count only process which using more then 99 % CPU ?

birkch wrote:

Welcome to the Forum @ankit360,

did you have a look to the binary behind the command "procs" ?

There is a example in the help:

Code

1. check_procs -w 10 -c 20 --metric=CPU
2. Alert if CPU of any processes over 10%% or 20%%

Example output:

Code

1. CPU OK: 129 processes | procs=129;;;0; procs_warn=0;;;0; procs_crit=0;;;0;

If you want to configure that in icinga2 you can do it like this:

Code

1. apply Service "procs" {
2. command_endpoint = host.vars.client_endpoint
3. check_command = "procs"
4. vars.procs_critical = 99
5. vars.procs_metric = "CPU"
6. assign where host.vars.client_endpoint
7. }

Display More

i tried it but its not working . BTW from where you got attribute ?

nothing such attribute is mention in icinga official document .

https://docs.icinga.com/icinga…n-check-command-processes

nilm wrote:

Code

1. object CheckCommand "procs" {
2. import "plugin-check-command"
7. command = [ PluginDir + "/check_procs" ]
12. arguments = {
13. "-w" = {
14. value = "$procs_warning$"
15. description = "Generate warning state if metric is outside this range"
16. }
17. "-c" = {
18. value = "$procs_critical$"
19. description = "Generate critical state if metric is outside this range"
20. }
21. "-m" = {
22. value = "$procs_metric$"
23. description = "Check thresholds against metric"
24. }
25. "-t" = {
26. value = "$procs_timeout$"
27. description = "Seconds before plugin times out"
28. }
29. "-T" = {
30. set_if = "$procs_traditional$"
31. description = "Filter own process the traditional way by PID instead of /proc/pid/exe"
32. }
33. "-s" = {
34. value = "$procs_state$"
35. description = "Only scan for processes that have one or more of the status flags you specify"
36. }
37. "-p" = {
38. value = "$procs_ppid$"
39. description = "Only scan for children of the parent process ID indicated"
40. }
41. "-z" = {
42. value = "$procs_vsz$"
43. description = "Only scan for processes with VSZ higher than indicated"
44. }
45. "-r" = {
46. value = "$procs_rss$"
47. description = "Only scan for processes with RSS higher than indicated"
48. }
49. "-P" = {
50. value = "$procs_pcpu$"
51. description = "Only scan for processes with PCPU higher than indicated"
52. }
53. "-u" = {
54. value = "$procs_user$"
55. description = "Only scan for processes with user name or ID indicated"
56. }
57. "-a" = {
58. value = "$procs_argument$"
59. description = "Only scan for processes with args that contain STRING"
60. }
61. "--ereg-argument-array" = {
62. value = "$procs_argument_regex$"
63. description = "Only scan for processes with args that contain the regex STRING"
64. }
65. "-C" = {
66. value = "$procs_command$"
67. description = "Only scan for exact matches of COMMAND (without path)"
68. }
69. "-k" = {
70. set_if = "$procs_nokthreads$"
71. description = "Only scan for non kernel threads"
72. }
73. }
78. vars.procs_traditional = false
79. vars.procs_nokthreads = false
80. vars.procs_warning = 250
81. vars.procs_critical = 400
82. }

Display More

Found in /usr/share/icinga2/include/command-plugins.conf

i even set vars.procs_pcpu = 99 in command-plugins.conf still no luck

i found that check is executed twice .

first check executed with command parameters which i specified in service configuration file and other check executed without command parameters .

any idea why it check it twice ?

below is debug.log