WG: Nagios schedule question

[h.baecker at ecofis.de]

Hi,

my last comment about the problem that I thought that Nagios is not able 
to start more than 60 events within one minute was not right.

It seems that our Nagios have problems with starting the count of events 
in a Event Loop that it should start.

I think (please correct me if I am wrong) that Nagios calculates the count 
of events per event loop out of the hole services and the average check 
interval.
The Nagios "scheduler" now have his event list that look semilar like 
this:

13:00:00        -       Start for example 50 Service Checks
13:00:01        -       Reaper + other service checks starts
13:00:02        -       Nagios could not start all the events out of the 
last loop and will do this now, means Nagios has now for example 50 
regular NOW checks + 20 out of the last loop...
 and so on, and so on...

out of this theory the latency for each service check will increase.

If I am right, how can I tune up my Nagios to start the correct number of 
Service Checks / Events for each loop?

Sorry for my bad english, I hope that someone understand me.

Hendrik


----- Forwarded by Hendrik Baecker/ECODOR/ECOFIS on 02.07.2004 15:32 -----

h.baecker at ecofis.de 
Sent by: nagios-users-admin at lists.sourceforge.net
29.06.2004 12:49

To
nagios-users at lists.sourceforge.net
cc

Subject
[Nagios-users] Nagios schedule question







Hi List, 

I have some problems with our Nagios and I think the problems comes from 
the scheduler. But before some Info: 

OS: Suse Linux 
Kernel: 2.4.[unknown] 
CPU:         2x Intel Intel(R) Xeon(TM) CPU 2.00GHz 
Mem:        2.587.208K 

Nagios Version 1.1 

Hosts:                616 
Services:        2684 

Average Service Check Intervall ~ 350 seconds. 

With the option inter_check_delay_method=s our nagios is not able to check 
all the services within 5 minutes (just 70%) and I can see out of the 
perfdata many latency about 100 up to 350 seconds. 

I tried to inter_check_delay_method=0.01 without any good result but an 
inter_check_delay_method=1.00 slows down nagios with his checks but 
doesn't throw out any latency. 

I think because of the unsigned long int for the time_t I guess that 
Nagios is not able to do more schedule events than 60 in 60 seconds. 

What can be done to screw up my Nagios Performance? Any ideas? 

Regards 
Hendrik 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.monitoring-lists.org/archive/users/attachments/20040702/a38c13aa/attachment.html>