New Nagios implementation proposal

[nap]

Hi,

Here is the result of a 100K conf with the random check (sleep random
from 0 to 20s if i'm not wrong) and an Merlindb export in the same
server. I will run with a 150K conf, but I think this shows we can
have far more perf if we bypass reaping and send results direct into
memory.



Jean

On Thu, Dec 3, 2009 at 10:36 AM, nap <naparuba at gmail.com> wrote:
> Thanks for this check, gcc in progress :)
>
>
> Jean
>
> On Thu, Dec 3, 2009 at 9:43 AM, Andreas Ericsson <ae at op5.se> wrote:
>> On 12/02/2009 02:45 PM, nap wrote:
>>> Thanks this is very usefull, I'll look at it :)
>>>
>>> I test with a random check 1->10 and it's ok for latency. It just
>>> needs a lot of workers so I'm trying to allow workers to launch
>>> several process at a time and so use less workers (from 200->4) and
>>> from now it's even faster an use less memory (one worker is a fork of
>>> poller, and poller is near 10Mo of resident memory so less Worker,
>>> less memory used). I just test with a 200K random time check, 4
>>> workers, 512 process/worker and it runs fine (load average at 3 and
>>> still some %idle available :) ).
>>>
>>> Now I'm trying a 300K configuration (still the same random time check) :D
>>>
>>
>> Try with the attached plugin. It's supposed to simulate a network under
>> extreme stress (ie, when monitoring is most important) by exiting with
>> a random exit-code.
>>
>>> But remember : Shinken is far more than good performances. Its
>>> architecture is also important and I would like to have your feed back
>>> about it too.
>>>
>>
>> I'll look into it when I get a few moments to spare. However, I'm fairly
>> certain Nagios has too much momentum and too many in-core modules for any
>> incompatible competitor to take its place.
>>
>> --
>> Andreas Ericsson                   andreas.ericsson at op5.se
>> OP5 AB                             www.op5.se
>> Tel: +46 8-230225                  Fax: +46 8-230231
>>
>> Considering the successes of the wars on alcohol, poverty, drugs and
>> terror, I think we should give some serious thought to declaring war
>> on peace.
>>
>> ------------------------------------------------------------------------------
>> Join us December 9, 2009 for the Red Hat Virtual Experience,
>> a free event focused on virtualization and cloud computing.
>> Attend in-depth sessions from your desk. Your couch. Anywhere.
>> http://p.sf.net/sfu/redhat-sfdev2dev
>> _______________________________________________
>> Nagios-devel mailing list
>> Nagios-devel at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/nagios-devel
>>
>>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 100K_random_mysql.png
Type: image/png
Size: 228517 bytes
Desc: not available
URL: <https://www.monitoring-lists.org/archive/developers/attachments/20091204/ac384e99/attachment.png>
-------------- next part --------------
------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
-------------- next part --------------
_______________________________________________
Nagios-devel mailing list
Nagios-devel at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-devel