aix 5.3 plugin to monitor for long running processes ?

Morris, Patrick patrick.morris at hp.com
Thu May 20 18:15:24 CEST 2010


Mirza Dedic wrote:
> Is there such a plugin to monitor aix topas utility (topas -P) and 
> report back processes that have a long running "TIME" value? Better 
> yet, TIME combined with high CPU value?
>  
> The output of topas -P on a AIX box looks like this:
>  
> _USER        PID    PPID PRI NI   RES   RES SPACE    TIME CPU%  I/O  
> OTH COMMAND_
> *mirde**   **4079860** **3244406** 120 20  1668  1865  1668  
> **20**6:26 2**5**.6**    0    0 **_progres*
> dougr   4079860 3244406 120 20  1668  1865  1668    6:26 20.6    0    
> 0 _progres
> prosvra 3412254 4071460 114 20  1453  1561  2033   16:09 18.7    0    
> 0 _proapsv
> braro   1654968       1  66 20  1388  1865  1793   57:25  1.5    0    
> 0 _progres
> batch   2613370       1  69 22  1689  1865  1879    7:05  1.2    2 
> 1220 _progres
> soual   1826826  286914  61 20  1806  1865  1806    0:16  0.8    0 
> 3376 _progres
> braro   1601808       1  61 20  1957  1865  2410   82:17  0.7    0    
> 6 _progres
> allbr   1274340 3399852  61 20  2061  1865  2061    0:01  0.5    0 
> 2312 _progres
> kelca   3407948 1413528  62 20  2207  1865  2207    0:08  0.5   42  
> 936 _progres
> allbr   2543698 1274340  62 20   208    61   208    0:00  0.4    0 
> 3320 ksh
> mikko   1032604 2073082  60 20  2522  1865  2522    0:15  0.4    2 
> 1255 _progres
> larha   3809412 2060626  61 20  2159  1865  2159    0:02  0.4   17  
> 115 _progres
> kirpa   1605690 3838314  60 20  2558  1865  2558    0:32  0.4    2   
> 13 _progres
> braro   2732514       1  60 20  1632  1865  2149   13:05  0.3    0    
> 0 _progres
> braro   1909064       1  60 20  1134  1865  1760    6:20  0.3    0    
> 0 _progres
> root    2719758  745934  58 41  2431   128  4619   24:25  0.3    0    
> 1 topas
> judre    549372 1503500  61 20  1798  1865  1798    0:00  0.2    1  
> 198 _progres
> prosvr  1814718       1  82 20  6374    26 15995    7:21  0.2    1    
> 1 java
> dorod   2904242 2187728  60 20  2495  1865  2495    6:53  0.2    1  
> 930 _progres
> linso   3760208 4395326  60 20  1509  1865  1509    0:00  0.1    0    
> 2 _progres
> katpa   3494134 4280722  60 20  2307  1865  2307    0:03  0.1    0    
> 0 _progre
>  
> What I want to capture is, something with 120+ (for TIME) and 20%+ (CPU).
>  
If you've got even basic scripting skills, Nagios plugins are incredibly 
easy to write.  Documentation is here:

http://nagiosplug.sourceforge.net/developer-guidelines.html

However, the most basic plugins only need to worry about a couple 
things: They should have one line of output, and return an exit code of 
0 for OK, 1 for warning, 2 for critical, and 3 for unknown.

Given that, all you'd need to do is write a simple script that runs that 
command and parses it looking for times over a certain value, and 
returns a 1 or a 2 if something crosses your thresholds.

------------------------------------------------------------------------------

_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null





More information about the Users mailing list