[Gluster-devel] FreeBSD port of GlusterFS racks up a lot of CPU usage

[Kaleb S. KEITHLEY]

On 12/30/2015 01:22 PM, Hubbard Jordan wrote:

> I also have a broader question to go with the specific one:  We 
> (at iXsystems) were attempting to engage with some of the Red Hat 
> folks back when the FreeBSD port was first done, in the hope of 
> getting it more “officially supported” for FreeBSD and perhaps even 
> donating some more serious stress-testing and integration work for 
> it, but when those Red Hat folks moved on we lost continuity and 
> the effort stalled.  Who at Red Hat would / could we work with in 
> getting this back on track?  We’d like to integrate glusterfs with 
> FreeNAS 10, and in fact have already done so but it’s still early 
> days and we’re not even really sure what we have yet.
> 

Hi,

To me, from a community standpoint, to be "officially supported" I'd
venture to say that what it takes is being visibly involved in the
project. That can take many forms, e.g., do regular builds on your
platform, submit bug reports (to our bugzilla) and associated fixes (to
our gerrit), implement and contribute new features, review other
people's patches in gerrit, build packages for your platform, evangelize
GlusterFS, answer questions in IRC and the mailing lists, etc., etc.

Everything that goes on the community is done by volunteers. There are
no Red Hat employees whose sole responsibility is to work on Community
GlusterFS. (Excepting our Community Manager, Amye.) The Red Hat mantra
is "upstream first" so every feature and every bug fix that Red Hat
employees work on does indeed go into Community GlusterFS first; a lot
does get done as a side effect of that policy, but nobody should take it
for granted that _everything_ (or anything) will just get done.

Nobody would say no to having serious stress testing and integration
work. If it plugs into our current gerrit and jenkins infrastructure, so
much the better. If there are people in your community who can help
maintain and/or grow our infrastructure, we could use a lot of help there.

With that level of involvement, I could imagine eventually FreeBSD
having more of a, I don't know, for lack of a better word, 'standing' in
the GlusterFS community. We do compile every patch on FreeBSD to ensure
that we don't break that level of portability, but that's the extent of
it. Maybe elevated to running regressions, as we do for NetBSD, which
has a bit of a legacy standing in the community due to Emmanuel Dreyfus'
long time participation.

Anyway, that's my opinion. (Emphasis on my and opinion. Perhaps others
will weigh in with their opinions.) I look forward to your involvement
in the community. Look for us at FOSDEM, a couple of us will be there.

-- 

Kaleb