Nagios is dead! Long live Icinga!

Andreas Ericsson ae at op5.se
Thu May 7 13:32:57 CEST 2009


Matthias Flacke wrote:
> Steven D. Morrey schrieb:
> 
>> What was wrong with Nagios from either a user or developer
>> perspective that you felt was not, or could not be addressed
>> through the community.
>> 
> 
> A second look onto users and developers perspective shows: during
> Nagios 3 development there was a good chance to get fixes and smaller
> changes into the development process. But after the release of
> Nagios3 the big black hole was opened, obvious errors e.g. in the
> scheduling (next check will be in 2010) and timeperiods remain
> unhandled. Until then we encouraged users to examine such errors and
> to provide detailed reports or even patches to the devel list. But
> that was more and more in vain, like Gerhard has described. So far
> concerning our personal motivation (speaking for the community
> members in our team) to bring some progress into nagios development.
> 

I agree with this. The issue is not technical but social. It's rather
easily remedied, but it requires Ethan to do some work, or allow
someone else to do work on his behalf.


>> What does Icinga bring to the table or plan to bring to the table,
>> to address these issues?
>> 
> 
> There are three steps planned until the first release: 1. adding of
> multiple patches from devel list and other sources, from this step
> the core source will be accessible via the usual channels.

I'm not sure exactly which patches you're talking about, but I'm all
agog to see them once the code repositories are released.

> 2. fixing
> and improving the DB part (called IDO), though it will be remain
> fully compatible to ndomod and the existing DB format.

This I don't understand though. Why not take this opportunity to
create a database schema that's easier to manage and understand?
We (op5) determined during a three-week period of testing that
the largest long-term problem with NDOUtils is its database schema,
which prevents it from scaling to huge installations. Since it's
easy to remedy in a way that also makes it simpler to understand
and use, why not take that opportunity now?

> 3. GUI -
> defining a API to control data flow between DB and GUI and add a
> flexible and extensible GUI framework.
> 

This, I think, will be rather wasted effort. op5 will release its
new GUI as beta before the nordic meet on nagios (4th of June, iirc).
Availability has always beaten quality in the OSS world, and we're
too close to done right now to decide to stop and wait to see what
icinga might produce in the area sometime in October. Our code is
available for download as a git repository at http://git.op5.org/git
If nothing else, you should at least have a look at it to see what
the competition looks like ;-)

>> Is Icinga a true fork, could patches be made available to those on
>> Nagios?  Or will Icinga be a distinct product with a new pedigree?
>> 
> 
> For the first phase which will at least last until the release 1.0 is
>  launched in October, we plan to be really conservative with the
> core. It's wrong to say, that we won't touch the core, because we
> plan to add numerous of the outstanding patches from the devel list.
> But we will strictly keep the nagios compatibility for all levels,
> i.e. plugins, transports (NRPE,NSCA), data storage,

Does this also mean object sizes (nebstruct_host_data et al) won't
change? If they do, no eventbroker modules that work with Nagios
will work with icinga, and that would be rather sad, tbh.

> 
>> Will migration to Icing be as simple as loading up a new binary,
>> i.e. will it strive to maintain full compatibility with Nagios, or
>> is it trying to break the mold?
>> 
> 
> I've already stated it before, but say it again clearly: there is no 
> break, Icinga will be fully Nagios compatible.  Furthermore: we
> expect that you can test it easily within your existing setups to get
> a feeling for it.
> 

I'll take this as "The ABI will also remain exactly the same" then,
which should mean noone has to re-compile their eventbroker modules
to make them work with icinga. Right?

>> Who is backing this project?  An organization i.e. business of some
>> kind, or something more akin to a users group?
>> 
> 
> We are both, a team consisting of community members and employees of
> a commercial company. We think that neither the community on its own
> nor a company without community support can master such a challenge.
> 

One out of two is not too shabby. The community could do it without the
company, but the company couldn't do it without the community. I know
that at least some of Ethan's grief against Netways is not unfounded
(and so does Hendrik Baecker I believe; IIRC he was standing next to
me and Ethan during last years Nagios Conference in Nuremberg when the
original source of grief was discussed. Ah well).

>> What would be the incentive if any for businesses who have invested
>> heavily in a Nagios based infrastructure to switch?
>> 
> 
> Although it is a bit early to answer such a question, since the
> roadmap is just at its beginning,

(ie, Icinga doesn't know).

> one point to mention: in terms of
> investment protection its vital to put Nagios core development on a
> broader base.
> 

Agreed, but I disagree about the methods used (though I understand
it; Ethan *has* been absent for too long), and I'm not too fond of
the fact that Netways *seems* to have paved the way for it, even
if that's not what actually happened. Bernd Erk (who is a very nice
guy indeed) is the project manager and also Netways' CTO. You can't
tell me that's 100% kosher and expect me to believe it without
doubting even a little.

Ah well. I'm sure it'll all work out alright in the end somehow
though.

-- 
Andreas Ericsson                   andreas.ericsson at op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

Register now for Nordic Meet on Nagios, June 3-4 in Stockholm
 http://nordicmeetonnagios.op5.org/

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.

------------------------------------------------------------------------------
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




More information about the Developers mailing list