Notification slow on Solaris - 27+ seconds spent in "set_all_macro_environment_vars(TRUE); "?

[Guy B. Purcell]

On Jan 14, 2008, at 8:24 AM, Steffen Poulsen wrote:

> Following up on previous debug, in rc1 we have inserted a few more  
> debug statements in base/utils.c:

[...debugging print statements & sample output...]

> What can we do to lower this pretty large amount of time spend on  
> setting the macro evn vars?
>
> We're already running with use_large_installation_tweaks=1, but  
> perhaps there are other helpful options touching the logic going on  
> in set_all_macro_environment_vars?
>
> We're at 3.0rc1, Solaris/SPARC, T1000.

Given this, you're running Solaris 10, which means you could use  
dtrace to find out in detail what's happening inside  
set_all_macro_environment_vars(), including exactly where the  
slowdown is.  I'm sure that info would be extremely helpful to the  
developers in solving the problem, assuming it's something in the  
code (anyone else see this on Solaris 10?).

Sorry I can't help more (personally, I'm running on Solaris 9 still,  
with only a handful of hosts, and don't see this behavior at all).   
Hmm, you might see if it's related to the size of your monitor base,  
too, by reconfiguring it to only check a dozen or so hosts & their  
services, keeping the rest of the config the same.

As far as HW goes, that box should be an ideal monitoring node-- 
capable of handling lots of simultaneous threads.

-Guy


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
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