The nagios community wants to keep its open soul

Andreas Ericsson ae at op5.se
Thu Feb 25 18:07:41 CET 2010


On 02/25/2010 05:18 PM, Frost, Mark {PBG} wrote:
> 
> 
>> -----Original Message-----
>> From: Marc Powell [mailto:marc at ena.com]
>> Sent: Thursday, February 25, 2010 10:54 AM
>> To: Nagios Developers List
>> Subject: Re: [Nagios-devel] The nagios community wants to keep its open soul
>>
>>
>> As far as the future of Nagios, it is Ethan's project, not ours; it always has been and>we are along for the ride. We can have hopes and dreams about what it can become and>Ethan may buy into them, or he may not. We can propose contributions and he may accept>them, or he may not. Regardless, we will all make our own decisions about whether this is>compatible with our own goals. Some will stay, some will move on. C'est la vie.
>>
>> --
>> Marc
> 
> I agree with Marc.  However, one thing that continues to puzzle me is what exactly
> the core developers do.

As one of those new core developers (the other being Ton Voon), I think I can
respond to that.

>  It was my understanding that when Ethan announced the
> addition of some core developers that there would be more changes to the core.  I
> thought that meant that those core developers could
> 
> - evolve the Nagios core on their own based on their own judgments, agreement of
> other core developers and perhaps input from the community as appropriate
> 

That's true. We still have our own social lives and dayjobs to take care of
though. Fortunately, my dayjob involves working on nagios-related things,
which is one of the things we core developers are supposed to do.

The second is to rapidly respond to severe errors and security issues. When
such a severe error last arose (the DST change), Ton worked ceaselessly
several days to find and implement the correct fix. This was all taken care
of in a few days.

When a security issue last arose, I produced patches for it and handled
communication with Mitre (who assigns CVE id's for such things) during the
two days it took me to locate and fix the problem to my own and numerous
worldwide security experts' satisfaction.

The third is to accept patches from community members. We do that too,
but see below for more information.

The fourth thing we do is communicate what's going on on this here list
and respond to emails such as yours.

The fifth thing we do is to hack up new features ourselves and maintain
the code. The problem with that is that both me and Ton are paid to work
on Nagios, which means our bosses don't generally devote a lot of time
to work on features that noone has requested.

I think that more or less exhausts our responsibilities. I agree that
we're lagging behind with "the fifth" above but the other duties are
being taken care of in a rather timely manner if you ask me. Maybe
that's just my take on it though. Other people are ofcourse entitled
to their own opinions.

> - accept patches as appropriate from the community and implement them
> 

We do that. Although not all patches are accepted due to them being
either downright wrong, making Nagios incompatible with eventbroker
modules and other things. We also do not comment on patches within
minutes. I've got other things to do than to anxiously scan this email
list every five minutes. I usually read it once or twice a day every
workday, which is enough to not let patches slide more than a couple
of days at the most. If that's not fast enough, I'm sure others could
(and do, which is much appreciated) comment on patches and make sure
they're polished enough for inclusion, or dropped without my own, Ton's
or Ethan's intervention.

You must understand that a large portion of the work we do is to make
sure the patches work as they're intended and doesn't break anything
else, which is a tricky thing to assure in Nagios, and that's the
reason why Ton is spending so much time writing tests for it.

> It was not my understanding that Ethan would continue to be the sole person who
> touched the core and made all decisions about the core.  That is that if Ethan
> doesn't do it, it doesn't get done.  So I'm confused.
> 

Ethan isn't the only one doing things. Most of the work has been done
by the community the past year. I've applied at $dayjob for more time
to work on Nagios, and it's been accepted that I do that once Ninja
is out the door (March 26, according to the schedule), with all the
changes required to Merlin to make that happen.

I'll start by cleaning up the code a bit and make sure it's more
easily testable. Unfortunately, many of the changes which would
make the code leaner, faster and more easily testable would break
the ABI, which in turn makes compiled eventbroker modules useless,
which in turn requires a minor version bump (3.2 -> 3.3).

Other areas could ofcourse be improved first, but it would mean
revisiting those areas again after the larger parts have already
been fixed, and that makes it less tempting. Ah well.

> Now, I totally appreciate and admire what I've seen getting done in terms of smaller
> patches and particularly the Merlin/Ninja work, but that's not really core work.
> I know that the work that people like Andreas do is funded by Op5 and I don't have a
> problem with that.  I don't quite know where the time and possibly money would come
> from to develop the core, but I agree it does need to happen.  I've seen many
> good messages from Andreas about things that need to be redone/reworked in the core
> and have wondered how they ever will get done.  It's probably more work than
> members of the community would take on themselves.  Perhaps I would have hoped
> that Nagios Enterprises might fund the core team to make those changes but maybe
> that's my imagination.
> 

Ofcourse it's not more work than the community can handle. The community
has already done wonderful things for Nagios and it will continue to do
so.

> So I'm still a very happy Nagios user and like the work I see happening, I'm just
> wishing there'd be more.
>

Nice to hear it. Makes hacking Nagios even more fun :)

-- 
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.

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev




More information about the Developers mailing list