Nagios and the community

In association with heise online

Galstad hopes he can fill the role of a community leader, but told The H that he'd already brought on a new member of staff with community experience who will help spread out the workload of handling the project internally. Over time, he sees himself less involved with the everyday code and more involved with the architecture of the Nagios project, selecting the good and bad ideas (and explaining why) or developing ways to evolve submitted patches into something better.
Ethan Galstad - President of Nagios Enterprises

Asked about alternative development models, such as the Red Hat/Fedora model where a development tree runs actively ahead of a productised version, Galstad says "it could be that we move to that model, but first what we want to do is concentrate on getting the current model working". He notes that it would require more manpower to manage the patches in the development version under the Fedora model, and more to bring those patches back into a stable product core. It could potentially create a bigger problem for Nagios if the model changed when issues remained with the current development model.

On the commercial use of Nagios, Galstad was happy to see so many companies make commercial use of the software. He feels that for many open source projects it is essential to have commercial outlet to fund development to offset the cost in free time of that projects developers. "It's hard to justify putting your time into 'free' when you have a family or you have a job which is demanding longer hours or your interests change. What do you do to ensure that that project still continues, especially if people find it useful and are still using it".

Galstad explained the issue; "We actually have a problem in the Nagios community that some people don't want to admit to or acknowledge, but I've seen first hand. These guys that are out there developing what I consider critical add-ons for Nagios are, in danger of leaving their projects for different reasons. It's not because they don't like the Nagios community or don't like their projects but they are in a position where, for example, some of them don't use Nagios in their daily job. They used to, but they don't any more. So, for instance, the NSclient++ add-on, which is a great Windows agent which everyone uses to monitor Windows, it's written by a guy who no longer uses Nagios on the job, so he's doing all this stuff in his free time. And he's said it's not only that he's donating his time; he has to buy the Microsoft development tools or find someone to donate them."

A practical solution described by Galstad was the situation of a Nagios developer who had recently lost his job. Though Nagios are unable to employ him, they are prepared to suggest add-ons which customers are looking for, and if developed, will enter into a revenue sharing deal with the developer on a commercialised version. "I'm not going to criticise anyone who wants to commercialise their stuff. I'm going to ask 'How can I help?".

On the more general question of the future governance of Nagios, Galstad was open to a range of ideas, from a Nagios foundation to a governance board. But the priority was for him to find "corporate financial sponsorship of the Nagios project, and what I mean by that is how do we get funds in to give to the person who's writing that Windows add-on we all want. It's in the best interests of companies who commercialise Nagios, us included, who should be stepping up to sponsor the project". It seems to be a chicken and egg problem for Galstad; the overhead in creating new structures for Nagios development needs sponsorship. "You first have to see if there is interest, and beyond interest, real action and real donations that are being committed and, once you reach that point, you can say 'OK' now we have the resources, it makes sense to look at doing something official and separate and with that overhead. Nobody is going to want that overhead if nobody is going to sponsor it".

Galstad's planning is still in the early stages; "I haven't yet floated the idea of sponsorships, but it'll be interesting to see what we hear back in the next month or two". In the interim, he sees it as his responsibility to protect the Nagios heritage and name while developing a plan for the future where Nagios projects are sponsored giving the coders time to devote to engineering a better, faster, cleaner and more flexible Nagios core and a wider range of plug-ins.

See also: