Nagios Performance

[Marc Powell]

On May 5, 2009, at 1:09 PM, Roberto Balestra wrote:

> Hi all,
> I have 10.000 check, 2300 host, 632 hostgroup 19xx host dependence, 7
> contacs and 2 contactgroup (no escaltion at the moment).
> I sent alarm by email. Check is perl based for snmp query and ping.
> My server is 4 processor Xeon 2.8GHz (32bit), 3GB RAM 5 HD 36GB SCSI
> (2xRAID1 System+3xRAID5 /usr & /var).
> OS is Fedora Core 9. Nagios is 1.4.1 (it is old I know) with mysql for
> comment/downtime/hostext/serviceext/status.

Similar configuration here (nagios instance kept around for historical  
reasons) --

-bash-2.05b$ time bin/nagios -v etc/nagios.cfg

Nagios 1.3
Copyright (c) 1999-2004 Ethan Galstad (nagios at nagios.org)
Last Modified: 11-17-2005
License: GPL

Reading configuration data...

Running pre-flight check on configuration data...

Checking services...
	Checked 8538 services.
Checking hosts...
	Checked 2784 hosts.
Checking host groups...
	Checked 469 host groups.
Checking contacts...
	Checked 16 contacts.
Checking contact groups...
	Checked 13 contact groups.
Checking service escalations...
	Checked 0 service escalations.
Checking host group escalations...
	Checked 0 host group escalations.
Checking service dependencies...
	Checked 0 service dependencies.
Checking host escalations...
	Checked 0 host escalations.
Checking host dependencies...
	Checked 0 host dependencies.
Checking commands...
	Checked 154 commands.
Checking time periods...
	Checked 5 time periods.
Checking for circular paths between hosts...
Checking for circular service execution dependencies...
Checking global event handlers...
Checking obsessive compulsive service processor command...
Checking misc settings...

Total Warnings: 0
Total Errors:   0

Things look okay - No serious problems were detected during the pre- 
flight check

real	0m28.007s
user	0m27.910s
sys	0m0.040s

>
>
> When I exec status.cgi (but is the same for any other cgi) it takes  
> more
> than one minute.

Well known issue with 1.x and large installations (at least it was  
well known when 1.x was widely used)

> When I check my configuration it takes more than one minute
> (real time by command 'time  /usr/local/nagios/bin/nagios -v
> /usr/local/nagios/etc/nagios.cfg')

Probably the structures referenced below and parsing of the host  
dependencies would be my guess.

> How can I improve my server's performance?

Upgrade. It's the single best thing you can do. You're also unlikely  
to find significant resources for 1.x support here any more. I don't  
believe there are many left that used it, let alone remember it's  
specific idiosyncrasies.

Some of the structures used in nagios-1.x  were not optimized for  
large installations such as ours. nagios-2 included object lookup  
enhancing changes that would help with the status.cgi times (search  
for 'nagios chained hash patch'). That patch was originally written  
for 1.2 so you could probably apply it to 1.4 if you can find it and  
you can't upgrade. I saw significant improvements with the patch and  
again by moving to nagios-2. I may have still have the patch but I  
can't say for certain if I modified it from it's original form (it's  
dated 2003 ;) ). If you can't find it I can send it to you,  
unsupported and YMMV.

nagios-3 offers further improvements for large installations. I can't  
vouch for it as I haven't moved to it yet but the enhancements look  
very positive.

--
Marc
  

------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image 
processing features enabled. http://p.sf.net/sfu/kodak-com
_______________________________________________
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