Is nagios core extendable? (fwd)

LadislavHavlik LadislavHavlik at seznam.cz
Tue Apr 7 17:01:09 CEST 2009


Sorry, my mistake.
Here is one row sample output:

NMP is tag (Nagios Mysql Procss), process is tablename, timestamp, hash_code, row size, host_name/crc, body.

NMP:process;1239115837;     24883;       323;skirit69.ics.muni.cz/2086;name/4:varchar(128):init;login/4:varchar(40):root;uid/1:int:0;realName/4:varchar(128):
root;pid/1:int:1;state/1:varchar(2):S;ppid/1:int:0;rss/3:int:424;vss/4:int:6120;path/20:varchar(255):variable uninitiated;cmdline/18:binary_code:init [2]..........;+;


--Lada

> ------------ Původní zpráva ------------
> Od: LadislavHavlik <LadislavHavlik at seznam.cz>
> Předmět: Re: [Nagios-devel] Is nagios core extendable? (fwd)
> Datum: 07.4.2009 16:54:55
> ----------------------------------------
> Hi Andreas,
> 
> well I must describe more information about solution:
> 
> in our concept is Nagios - only part; "collector" for a lot of values, which are
> 
> send (as passive tests, or active test) from each node = we are distinguish 
> between sensors and tests(test poduces only state and "easy" plain text. 
> Sensor produces more structured information that can describe any
> mysql table(s) (it is META data and data selfs), I may append one row =
> output from sensor, that contain "pre-evaluation":
> It could be good.
> 
> Yes, I don't need defined mysql tables (in advance time), each sensor defines
> table 
> self and Nagios can create table if not exist and saves data into tables... One
> note:
> first is update if operation is "failed" then insert.
> 
> One format in structured format is binary_code.
> 
> Yes, but event broker module does solve problems such is cumulative logs (we
> have 
> service data = meta data and "data". Where I can read more about event broker
> developing
> (I have only "halloworld module", but I need more info, about modules design,
> how connect
> to nagios, possibilities...more :))?
> Well, I can go back, from hard source code programming and modify to module
> programming, 
> I think that is probably better solution (maintainability).
> 
> 
> Thanks 
> Lada
> 
> Sample one row sensor output code:
> 
> 
> 
> > You can do that, provided the resulting fork is still licensed
> > under GPL, the same as Nagios is today.
> 
> > > 1. direct saving results into DB (not by using NDO etc..., but direct -
> ASAP
> > but refllex possibilities
> > > DB and work node ~ such is latency, IOWait, busy Waiting, timeouts,
> etc...),
> > >
> > 
> > That would be stupid for a lot of reasons, not least of which is
> > that you'll have to maintain it separately forever and ever.
> > 
> > > 2. "predictive - modul", but no stnd. Nagios modul (this isn't
> > > good place for more explanation).
> > 
> > You're wrong. This is exactly the right place for more
> > explanation.
> > 
> > > Very short: Not concept "Event done then send message" but
> > predicitve-statistically: "EVENT PROBABLY
> > > is able to occur then send message".
> > >
> > 
> > You can get at this information without modifying the Nagios
> > core. If you still want this to run in the core of Nagios, a
> > far better way of achieving that is to write an eventbroker
> > module.
> > 
> > >
> > > Have I got your suport for this solutions?
> > 
> > That depends on what you mean by our support. I won't write
> > it for you, but if it turns out to be a nifty piece of code
> > I might use it for something and then ofcourse I'll send you
> > bugreports and patches.
> > 
> > > Are these solutions correct?
> > 
> > Since you haven't created the solutions yet, that's impossible
> > to tell.
> > 
> > > What is your opinions on this?
> > 
> > That you should refrain from altering the Nagios core if you
> > can manage by getting the required data from other places,
> > such as by loading a module or writing a log-parsing program.
> > 
> > > Thanks for reply
> > >
> > 
> > You're welcome.
> > 
> > >
> > > Note:
> > > Sorry, but I have very good reason for this short info - no detail for more
> > info.
> > >
> > 
> > That sentence doesn't really parse. Anyway, you'll find it hard to get help
> > from the community while you're withholding details about your project.
> > 
> > 
> > -- 
> > 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.
> > 
> >
> ------------------------------------------------------------------------------
> > This SF.net email is sponsored by:
> > High Quality Requirements in a Collaborative Environment.
> > Download a free trial of Rational Requirements Composer Now!
> > http://p.sf.net/sfu/www-ibm-com
> > _______________________________________________
> > Nagios-devel mailing list
> > Nagios-devel at lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/nagios-devel
> > 
> > 
> > 
> 
> 

------------------------------------------------------------------------------
This SF.net email is sponsored by:
High Quality Requirements in a Collaborative Environment.
Download a free trial of Rational Requirements Composer Now!
http://p.sf.net/sfu/www-ibm-com




More information about the Developers mailing list