Up to this point, all of the rolekit work has been done ad-hoc
(primarily between Thomas Woerner and myself. I am pleased to say that
in the last week, rolekit involvement has grown significantly.
First up, Alex G.S. who has contacted me directly and will be working
on refactoring rolekit's systemd unit creation to use the python
-systemdunit library (which should hide some of the ugly
implementation details from role creators).
Second, we have Nils Philippsen who stormed into the scene this very
morning with his very first patch contribution. Thanks, Nils! I look
forward to many more.
With the team now starting to grow, I think it's time for us to
schedule a weekly status meeting and bug/rfe triage. I've created a
WhenIsGood request that I'd like anyone interested in rolekit
development to respond to so that we can schedule this. We will hold
these meetings in public (probably on one of the #fedora-meeting
channels so we can have them logged by zodbot).
I don't know if it's a basic thing to use virt-customize or guestfish
on Fedora Server, but as soon as I started to dive into using Fedora
Cloud images as containers I needed to use both, but had to install
them first. Not a big deal, but I wonder if integration of products
improves if libguestfs-tools is already available on Server?
I only have one thing on my agenda for today:
* Status update on memcached server role
That was pretty much covered in the earlier email I sent out requesting
review help, so if people want to just skip the meeting this week, I'm
At this week's Fedora Server SIG meeting, I indicated that things were
looking bleak for landing the containerized memcached role in F23. It
turns out things weren't as bad as I feared (with some caveats).
I was able to find the time this week (mostly by turning off IRC) to
get the memcached role built. It's not the final version I was hoping
for (producing a nulecule app), but it *can* serve memcached using the
official Fedora Dockerfiles version of the service uploaded to the
What do I need help with?
* I need someone to run through the open code review requests. I
think things are in good shape, but I would prefer a second set of eyes
So, what can it do right now?
* You can deploy a memcached role entirely on the local machine, the
same way that you can deploy the database server role or the domain
* You can deploy the role with no arguments (it has sensible
* You can pass arguments to control the cache size, number of
connections and number of threads.
* You can pass a special boolean argument "contained". This will cause
the role to be deployed by pulling down the fedora/memcached image from
Docker Hub and creating a systemd unit file to load it at boot.
* The containerized version is just that; memcached in a container. It
has no advanced orchestration capabilities.
* The version of the fedora/memcached image in the Docker Hub doesn't
have the capability to tune the same values as the local version. Right
now, these are ignored. I have a pull request open to fix this.
On Fri, Jul 10, 2015 at 04:51:15PM +0100, Joe Brockmeier wrote:
> > provided elsewhere that let you do that already. Conversely, atomic
> > is compelling _because_ of the Atomic platform. Atomic has novelty
> > (for now), decent technical advantages, and a lot more marketing
> > behind it.
> Well, I mean... it's compelling for us because we want people to have
> Fedora available $all_the_places, right?
> So having only Atomic means we'd basically be saying if you want to do
> things in the cloud, either do them the "Atomic way" or use another
> project, right?
So, another option here is to twist an early decision we made by 90°.
Switch from Cloud, Server, Workstation to Atomic, Server, and
Workstation (sorry, branding/design team!). Atomic can run on bare
metal, and we could start making Fedora Server cloud image and push
those to Amazon, etc., and this could serve as the generic
building-block cloud image people want. It'd probably be bigger, but
that's okay because we'd know _why_. (And I hope with the improvements
we're making on size in general, it wouldn't be too much bigger.)
Fedora Project Leader
#fedora-meeting-1: Server SIG Weekly Meeting (2015-07-14)
Meeting started by sgallagh at 14:59:44 UTC. The full logs are
* roll call (sgallagh, 14:59:44)
* Agenda (sgallagh, 15:03:44)
* Agenda Item: Getty IP Address (sgallagh, 15:04:19)
* Agenda Item: Fedora 23 Progress (sgallagh, 15:04:29)
* Agenda Item: Atomic Server (sgallagh, 15:04:39)
* Getty IP Address (sgallagh, 15:07:31)
* IDEA: Use the approach described in
with the \S amendment (sgallagh, 15:11:23)
* AGREED: Use the approach described in
with the \S amendment (sgallagh, 15:16:26)
* Fedora 23 Progress (sgallagh, 15:16:31)
* ACTION: sgallagh to file a ticket for a Server Role Test Day
* Fedora Server 23 will have few user-visible changes. Encourage
marketing to focus efforts on the other Editions. (sgallagh,
* Atomic Server (sgallagh, 15:33:59)
* Help wanted to investigate an Atomic Server platform. (sgallagh,
* Open Floor (sgallagh, 15:51:06)
Meeting ended at 15:57:32 UTC.
* sgallagh to file a ticket for a Server Role Test Day
Action Items, by person
* sgallagh to file a ticket for a Server Role Test Day
People Present (lines said)
* sgallagh (81)
* adamw (22)
* danofsatx (14)
* mitr (12)
* nirik (10)
* zodbot (4)
* jsmith (2)
* crose (1)
* mizmo (1)
* stefw (0)
* tuanta (0)
* simo (0)
Generated by `MeetBot`_ 0.1.4
.. _`MeetBot`: http://wiki.debian.org/MeetBot
Browse: No working groups are listed, including Server.
Search field: typing in "server" brings up two results: server-wg and
fedora-server, neither of which have any 2015 entries.
So, where are the meeting minutes? Since the new meetbot, I'm having
difficulty finding things, but server stuff I can't find at all.
Hi server group,
based on , the working groups are expected to gather feedback from
users that should help defining next steps in Fedora.next development.
Since there are no direct users of Env & Stacks group and it makes sense
to be able to segment the answers by audience, Env & Stacks group
decided to let other groups to ask their users some particular questions
, that are connected to Env & Stacks work, if it would be possible.
Thus, I'd like to find out whether there are some ideas already how to
gather feedback from your users or whether you plan to do something
where we can participate.