[Gluster-devel] Glusterd: A New Hope

Stephan von Krawczynski skraw at ithnet.com
Sun Mar 24 08:43:21 UTC 2013

On Sat, 23 Mar 2013 18:44:06 -0400
Jeff Darcy <jdarcy at redhat.com> wrote:

> On 03/22/2013 06:08 PM, Stephan von Krawczynski wrote:
> > Why is it you cannot accept that it should be a _filesystem_, and nothing else?
> > It would have been a lot better to care about stability, keep it simple and
> > feel fine. Concentrate on the strength (client based replication setups) and
> > forget the rest.
> "Just a filesystem" has historically been an obstacle to deployment of
> distributed filesystems, and just doesn't cut it any more.  It's
> important to have a coherent notion of which servers are up and which
> protocol versions they can accept.  It's essential for configuration
> changes to be coordinated and communicated across the cluster, if those
> changes are to be non-disruptive, and that's part of glusterd's job.  It
> also handles process management (both regular brick daemons and
> maintenance-related tasks), quorum enforcement, and other functions.
> The trend is for distributed systems to become more autonomous, not less so.
> If you want to run things in a 2.x fashion, feel free.  Volfiles still
> work, and will continue to do so, though you'll be giving up a lot of
> functionality that way.  Nobody else is asking us to turn back the clock
> and throw away functionality.  Whatever the problems might be with
> glusterd's implementation, the solutions lie ahead of us.  What's behind
> us should and will still that way.

Your current path is as wrong as bringing up an all-in-one solution, remember
the gluster distro? It was the same story and vanished for very good reasons.
Your problem with complexity is not distribution of vol-files. Read the list,
almost nobody ever complained about that. Instead the list is full of people
searching for the best (i.e. optimized) _content_ of the volfiles for their
respective setup. So the automated distribution of the vol-files does not help
the majority at all.
I accept there may be a small minority for this, and maybe they paid for it.
But the main problem with glusterfs is that it does not detect the best setup
of translators _by itself_. So iff you want to really help people create a
vol-file-creator that spits out the best possible configuration for a
user-defined setup by asking _few_ _easy_ user questions.


More information about the Gluster-devel mailing list