Tuning Question
Mark Ahlstrom
mark.ahlstrom at managedmail.com
Thu Mar 17 17:50:38 CET 2005
As an update, when it comes to tuning Latency I've found a couple of
interesting things. And I wish to share them.
First, Google really misses the mark on Nagios searches. Once I started
looking through Yahoo I found some really good leads.
I found a reference to ./nagios -s ../etc/nagios.cfg . And I started
tweaking. Searching through this list I found the most useful hint: it's
a matter of tweaking both max_concurrent_checks and the
service_reaper_frequency. If the reaper is too low, you could have a
problem. So I started with the minimum recommended max_concurrent_checks
from "nagios -s" and started tweaking my reaper from 5 downward. Once my
Latency average was almost better, I started upping my
max_concurrent_checks until I was comfortable.
Right now my values are: max_concurrent_checks = 100 and
service_reaper_frequency = 3, with 3502 services and 280 hosts.
NOW FOR THE INTERESTING PART! From "nagios -s" there is a nice warning,
"The minimum value also reflects best case scenarios where there are no
problems on your network." I'm building a fail over pair on the 280R's.
And the only difference between the V210 and the 280R's (other than
hardware) was the network. As a controlled test I placed one of the
280's on the same network as my 210, and synced the configuration files
on both 280's. The results are dramatic. The average Latency for my new
network is 275 seconds. The average Latency for the old network (where
the 210 resides) is .788 seconds! Needless to say, I'm going to pull out
the cat o' five tails and "visit" out network admin. 8-)
I hope this helps someone later on.
On Mon, 2005-03-14 at 17:57 -0600, Mark Ahlstrom wrote:
> I have a tuning question for Nagios 1.2.
>
> I'm currently migrating my core nagios server from a SUN V210 to a SUN
> V280R and I'm finding the check latency to be worse on the 280 than the
> 210 (the opposite of what I expected). Whenever I force a check through
> the web interface the latency factor is about 60 seconds on the 210 and
> about 300 seconds for the 280. Both servers were built the same way with
> the same configurations (OS and Nag).
>
> Is there some way I can speed up the scheduling queue? I have already
> set the reaper frequency to 1 and max concurrent checks to 900 with no
> noticeable result.
>
> I'm checking 3500 services and 280 "hosts" (some are virtuals).
>
> Another question would be, am I asking a moot question because of my
> platform choice? Because of GCC and compile options would I be better
> off with a Linux based server?
>
> Thanks,
> Mark
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
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