nagios future?

[Andreas Ericsson]

Michael Friedrich wrote:
> Hi,
> 
> Andreas Ericsson wrote the following on 09.06.2009 11:10:
>> Yes. It was discussed quite a lot some few weeks back on the
>> nagios-devel mailing list. Browse the archives for the full
>> discussion.
>>   
> try 
> http://sourceforge.net/mailarchive/message.php?msg_id=E03A84B43BAE443888372020DDBFE0E7%40int.consol.de 
> 
> 
> If you're interested in reading a bit more, try
> 
> http://sourceforge.net/mailarchive/forum.php?forum_name=icinga-users
> http://sourceforge.net/mailarchive/forum.php?forum_name=icinga-devel
> 

Thanks for those links. I'm far too lazy to look them up myself ;-)

>>
>> The future of Nagios is looking quite bright. In all honesty, that is
>> in part thanks to the Icinga fork, which has sparked a flurry of activity
>> within the Nagios developer community.
>>   
> And also popping up many community based sites, beside the existing 
> ones. Not that bad, but a bit misleading for new users imho. But let's 
> see how it resolves in a bit.
> Hopefully Nagios will be on GIT soon to merge knowledge from both 
> projects together. Dunno what plans are going on concerning the NDO and 
> other similar core parts but I think there's much potential to share 
> ideas and kniowledge between Nagios and Icinga.

Nagios will move to git when 3.2.0 is out the door. Ethan wants some
time to manage patches and stuff like he's used to without having to
learn another tool. I'm sure he'll curse himself for not switching
sooner when he learns the benefits of git, but at least we're getting
there.

One of the annoying things about the icinga-fork though is that they've
mainly done a lot of renaming and not so much actual patching. This will
ofcourse merge cleanly but in an unsatisfactory way for Nagios. Messy,
but certainly possible to work around.

>> First of all, we'll be releasing 3.1.1 soon, containing a plethora of
>> bug- and performance fixes. Ethan's working on automating the release
>> process so that Ton and I can cut releases without having to update a
>> bunch of webpages, sourceforge downloads area, documentation, etc, etc.
>> 3.1.1 will be the first live test of that automated process. If it drags
>> out another week or so though, we'll probably just go ahead and do it
>> manually anyway, as 3.1.1 really has a lot of important fixes that the
>> Nagios users really should get their hands on.
>>   
> It would be great to mention that all even releases are stable while odd 
> remains testing. On nagios.org 3.1.0 is only mentioned as "latest 
> version" and after clicking the download link it is marked as testing - 
> bit confusing, but not really a problem for experienced users.

Oh, right. I'd actually forgotten that.

>> Nagios will get a new GUI, dubbed "Ninja" sometime during or after the
>> summer. Ninja is available for download already and is "usable but has
>> some warts and is still incomplete" according to Ninja maintainer Per
>> Åsberg. You can find out more about it at
>> http://www.op5.org/community/projects/ninja. This was announced at the
>> Nordic Meet on Nagios which was held in Stockholm just last week. Note
>> that it's not necessarily easy to install yet as it's still a work in
>> progress. Bug-reports or enhancement requests are ofcourse very welcome,
>> and documentation patches for the installation procedures even more so.
>>   
> By announcing Ninja as new GUI, the rumors get into Merlin for DB usage. 
> I've read several posts about that but my question is, how far would 
> that be realistic?


Very realistic. We're already using it for development to that purpose,
and it's working just fine. One problem with NDOUtils is that the database
schema makes it impossible to write stuff for it that scale linearly.
That's totally unacceptable for us, so we had to come up with something
new. Fortunately, Lars Hjemli of the NagVis project has been very friendly
and cooperative in helping us add support for the Merlin database schema
in NagVis. Given how simple the Merlin schema is, I have no doubt that
we'll provide patches to other projects to achieve the same thing.

> For what I know Merlin uses the libdbi (just as 
> modified IDOUtils for Icinga) so it would be possible to use different 
> db types. Are there any plans to realize that? :-)
> 

It's been planned, implemented, tested and available since 2009-03-17.
Additional bugfixes happened later, but libdbi has been in use in
Merlin almost three months now.

I'm working (but very slowly) on some patches to address the multiple
memory allocations required to use libdbi for quoting strings etc,
since it prevents us from using a static arena to do the quoting etc
in, but that will take a while to complete so we're living with that
microscopic deficiency for now.

$ git show 084cdc85
commit 084cdc85d7b0c8a4f721804476979e904e4afe7a
Author: Andreas Ericsson <ae at op5.se>
Date:   Tue Mar 17 10:44:47 2009 +0100

    Use libdbi for database abstraction
    
    In some ways it's worse, since we're now forced to allocate
    and deallocate a lot of memory for each request, but in other
    ways it's pure win as we can now let users use whatever
    database type they want.
    
    Signed-off-by: Andreas Ericsson <ae at op5.se>

-- 
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.

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
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