Execution Queue and latency
Bishop, Dean
dean.bishop at tcdsb.org
Tue Oct 29 16:48:17 CET 2002
Good morning,
Marc:
Hmmm, yes, you are quite right. i am not sure how to account for that and i
also have experience with nagios -s that tells me that it cannot be used for
reliable capacity planning. Mine tells me 17 minutes when 5 can do the job
quite nicely.
best of two evils??
later,
dean
-----Original Message-----
From: Marc Powell [mailto:mpowell at ena.com]
Sent: Tuesday, October 29, 2002 10:42 AM
To: Bishop, Dean; louafi; nagios-users at lists.sourceforge.net
Subject: RE: RE : [Nagios-users] Execution Queue and latency
Your estimation presumes that service checks are not being executed in
parallel, which is not the case for a default installation.
Sebastien:
You may want to look at your max_concurrent_checks value to make
sure it is sized properly for your installation. You can run 'nagios -s
/path/to/nagios.cfg' to get a recommendation ala
[root at noctools nagios]# bin/nagios -s etc/nagios.cfg
Nagios 1.0b3
Copyright (c) 1999-2002 Ethan Galstad (nagios at nagios.org)
Last Modified: 06-09-2002
License: GPL
SERVICE SCHEDULING INFORMATION
-------------------------------
Total services: 2030
Total hosts: 1677
Check reaper interval: 10 sec
Inter-check delay method: SMART
Average check interval: 300.000 sec
Inter-check delay: 0.148 sec
Interleave factor method: SMART
Average services per host: 1.210
Service interleave factor: 2
Initial service check scheduling info:
--------------------------------------
First scheduled check: 1035905875 -> Tue Oct 29 09:37:55 2002
Last scheduled check: 1035906175 -> Tue Oct 29 09:42:55 2002
Rough guidelines for max_concurrent_checks value:
-------------------------------------------------
Absolute minimum value: 68
Recommend value: 204
Notes:
The recommendations for the max_concurrent_checks value
assume that the average execution time for service
checks is less than the service check reaper interval.
The minimum value also reflects best case scenarios
where there are no problems on your network. You will
have to tweak this value as necessary after testing.
High latency values for checks are often indicative of
the max_concurrent_checks value being set too low and/or
the service_reaper_frequency being set too high.
It is important to note that the values displayed above
do not reflect current performance information for any
Nagios process that may currently be running. They are
provided solely to project expected and recommended
values based on the current data in the config files.
--
Marc
> -----Original Message-----
> From: Bishop, Dean [mailto:dean.bishop at tcdsb.org]
> Sent: Tuesday, October 29, 2002 9:28 AM
> To: 'louafi'; nagios-users at lists.sourceforge.net
> Cc: Bishop, Dean
> Subject: RE: RE : [Nagios-users] Execution Queue and latency
>
> well, assuming that you haven't modified "check_ping" it uses 4 pings to
> test the OK'ness.
>
> On my system it takes just over 4seconds to check_ping a host on the LAN.
>
> 4 seconds * 70 hosts = 280 seconds or just under 5 minutes.
>
> so it will take about 5 minutes just to do the check_pings if things are
> OK.
> it just gets worse if things are not OK because then each of the four
> pings
> would have to timeout (defaults to 10seconds).
>
> 2 suggestions
>
> 1) use:
>
> command_line /opt/nagios/libexec/check_ping -H $HOSTADDRESS$ -w
> $ARG1$ -c $ARG2$ -p 1 -t 5
>
> 2) use:
>
> command_line /opt/nagios/libexec/check_tcp -H
> $HOSTADDRESS$ -p 23 -w $ARG1$ -c $ARG2$
>
> substitute the port 23 for something useful. i use 23 for my switches and
> 135 for my windows boxes.
>
> btw
>
> check_tcp -h host -p 135 takes about .01s
> check_ping -h host -p 1 takes about .01s
>
> take your pick.
>
>
>
> later,
> dean
>
>
>
> -----Original Message-----
> From: louafi [mailto:sebastien.louafi at e-qual.fr]
> Sent: Tuesday, October 29, 2002 10:19 AM
> To: nagios-users at lists.sourceforge.net
> Cc: 'Bishop, Dean'
> Subject: RE : [Nagios-users] Execution Queue and latency
>
>
> Ok my check_ping definition is :
>
> define command{
> command_name check_ping
> command_line /opt/nagios/libexec/check_ping -H $HOSTADDRESS$
> -w $ARG1$ -c $ARG2$
> }
>
> I have around 70 hosts and 800 services...
>
> I'm using perfdata to dump check result in a mysql database.
>
>
> Thanks
>
>
> -----Message d'origine-----
> De : Bishop, Dean [mailto:dean.bishop at tcdsb.org]
> Envoyé : mardi 29 octobre 2002 15:59
> À : 'louafi'; nagios-users at lists.sourceforge.net
> Cc : Bishop, Dean
> Objet : RE: RE : [Nagios-users] Execution Queue and latency
>
>
> please send the snippet from checkcommands.cfg where "check_ping" is
> defined.
>
> also, how many hosts/services are you checking?
>
> -----Original Message-----
> From: louafi [mailto:sebastien.louafi at e-qual.fr]
> Sent: Tuesday, October 29, 2002 9:37 AM
> To: nagios-users at lists.sourceforge.net
> Cc: 'Bishop, Dean'
> Subject: RE : [Nagios-users] Execution Queue and latency
>
>
> OK i give you my config files.
>
> My services.cfg contains a lot of services like :
>
> define service{
> host_name 10.10.10.10
> service_description PING
> check_command check_ping!3000,20%!5000,60%
> max_check_attempts 5
> normal_check_interval 5
> retry_check_interval 3
> active_checks_enabled 1
> passive_checks_enabled 0
> check_period 24x7
> parallelize_check 1
> check_freshness 0
> flap_detection_enabled 0
> process_perf_data 1
> retain_status_information 1
> retain_nonstatus_information 1
> notification_interval 60
> notification_period 24x7
> notification_options w,c,r
> contact_groups louafi
> stalking_options o,w,u,c
> }
>
> My hosts.cfg is :
>
> define host{
> host_name 10.10.10.10
> alias 10.10.10.10
> address 10.10.10.10
> check_command check_alive
> max_check_attempts 5
> checks_enabled 1
> flap_detection_enabled 0
> process_perf_data 1
> retain_status_information 1
> retain_nonstatus_information 1
> notification_interval 60
> notification_period 24x7
> notification_options d,u,r
> }
>
> Thanks for your help
>
>
> -----Message d'origine-----
> De : Bishop, Dean [mailto:dean.bishop at tcdsb.org]
> Envoyé : mardi 29 octobre 2002 14:24
> À : 'louafi'; nagios-users at lists.sourceforge.net
> Objet : RE: [Nagios-users] Execution Queue and latency
>
>
> It would help if you sent a snippet from your services.cfg and hosts.cfg
>
> -----Original Message-----
> From: louafi [mailto:sebastien.louafi at e-qual.fr]
> Sent: Tuesday, October 29, 2002 5:18 AM
> To: nagios-users at lists.sourceforge.net
> Subject: [Nagios-users] Execution Queue and latency
>
>
> Hi,
>
> I have posted this problem few weeks ago without answers...
>
> I have a high latency value for my check executions.
>
> How to solve this problem ???
>
> I have the following options set :
>
> inter_check_delay_method=s
>
> service_interleave_factor=s
>
> max_concurrent_checks=0
>
> service_reaper_frequency=10
>
> Thanks a lot.
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
More information about the Users
mailing list