-------- Original Message --------
Subject: [ANNOUNCE] Deltacloud Community Call
Date: Fri, 09 Nov 2012 14:53:17 +0200
From: marios(a)redhat.com <mandreou(a)redhat.com>
We'll be having the first Deltacloud Community call next Thursday, 15th
November @ 1700 CET (Prague, Brno, GMT+1) - which means 0800 PST (US
West Coast, GMT-8) or 1100 EST (US East Coast, GMT-5).
The idea is to have two topics for discussion during a 1 hour call.
Please vote for topics on the etherpad @
http://piratepad.net/DeltacloudCall - by putting a +1(name) next to the
topic you like. Feel free to add your own topic - e.g.:
* Deltacloud clients - working with cURL, the ruby client, the web
browser (e.g. using a RESTClient plugin) - VOTES: +1(marios),
* How to file bug reports for Deltacloud - process, what to report,
following up. - VOTES +1(anna), +1(james)
The call in details are on the etherpad - I will arrange an elluminate
session if necessary (depending on topics) and send a reminder together
with the agenda before the call.
All are welcome!
On 02/11/12 14:50, marios(a)redhat.com wrote:
> we think it might be useful to have a weekly open/community call to
> discuss all things Deltacloud. The proposal is for 1 hour calls, the
> format of which is up for discussion. The idea right now is to have 2
> 'topics' per call; topics can be:
> 1. A presentation on a particular aspect - 'How do exceptions work in
> 2. A question about something you're working on - 'JIRA ticket
> DTACLOUD_555 - discussion and possible approaches'
> 3. An 'open' Q/A session with the devs.
> 4. Anything else you want/think might be useful...
> If on a given week there are no topics then we can just have an open
> discussion on "what i'm working on and how I've implemented it" for
> early feedback for example.
> So, as we are all all over the place - it may be tricky to find a
> suitable time. An entirely unscientific poll of a few people has
> indicated that the 0800 Pacific time may be a good slot... making it
> 1600 in Europe and 1100 in NYC. It also looks like Thurdsay/Friday would
> be a good day due to conflicts earlier in the week. Please indicate your
> preference on the poll @ http://doodle.com/c2ht9zf4e7gngrhu (be sure
> to change the time zone to whatever is local to you). If you can't make
> *any* of those times then just vote no to everything and reply to this
> mail with suggestions.
> A preliminary etherpad for the first call is @
> Comments and suggestions pretty please?
On 11/08/2012 02:51 AM, Miroslav Suchý wrote:
>> I would like to start pushing updates
>> to rawhide soon (rawhide only). To assist in the update process, I've
>> whipped up a little script (which you can find here ) called the Gem
>> Rpm Updater, or gru.rb
> `git reset HEAD~ --hard`
My thoughts behind this were to ensure a clean run every time this
script is invoked.
I'll probably incorporate the standard ruby option parser at some point
for more options and can add a flag such as -f to toggle overriding
local changes before attempting to make updates.
The following series of patches reorganizes the website
to simplify it and consolidate information. I tossed the
statically generated site up on one of my servers for
Included in this patchset:
- simplification of content, removing redundant items
- news removed, driven by the blogroll rss feed
- focused on the Aeolus being an umbrella project with
project specific resources managed by those communities
- embed the google map into the team section
- various other changes
Thoughts? Comments? ACKs?
On 29/10/2012, at 2:58 PM, Michael Orazi wrote:
> Looks good! I didn't see any mention of the other older repos. Are they completely shutdown at this point/obviously not taking pull requests, etc.?
> Thanks for writing this up, and great work across the board to make this happen.
(note - added aeolus-devel on CC to this)
During the Community talk this morning, it was mentioned there are a bunch
of pages that aren't being found through Google. But these same pages work
when a person uses the non "www." version of the website, because our redirects
take care of it.
As a thought, GitHub lets us do custom 404 pages:
to the correct end URL.
Does that sound workable?
Regards and best wishes,
Aeolus Community Manager
Hopefully I got the important stuff right :)
* Motivation - converge Aeolus Conductor and Katello web UI look, also converge
Conductor within itself
* MIT license
* Components - shell (header, footer, ...) vs. content (forms, ...)
* Currently git submodule
* More content components
* Developer guide
* Make it a separate open-source project
* Carl, Bert, the infrastructure consumers
* Dennis, the app designer
* Sam, the system administrator
* /Persona descriptions will be available later/
* A Rails App mountable into another
* Isolation of Ruby namespace, routes, views
* Uses the same DB as the host app, comes with own migrations
* DB tables are also namespaced
* Testing with a dummy app located in test/dummy
* Rails Engine that talks to ImageFactory
* ImageWarehouse will be dropped
* Tim works with ImageFactory v2
* Build is no longer there (it's a Conductor concept)
* Metadata also stored in Conductor
* ImageFactory's model: BaseImage ---< TargetImage ---< ProviderImage
* Tim responsibilities
* Template store
* Custom data
* Image imports
* Error & status reporting
* Metadata store
* Rails Engine for permissions
* The functionality is already present in conductor, now it's being factored out
* Reusable roles
* Assigned not to only user/group, but to pairs - (user, object) or (group,
object) => that results in (user, role, object) tuple
* E.g. Frank is an owner of deployable Abcdef
* Querying - `check_privilege` vs. asserting `require_privilege`
* Filtering ActiveRecord queries - `list_for_user`
* Requires classes to represent users and groups (class names are
configurable). The classes that represent target objects must include
as per subject, I'm currently looking at a scenario where one configures
ovirt/rhev with a single datacenter and multiple data domains
currently I don't think imgfac allows for configuration of the data
domain where the templates are imported
ovirt/rhev can only instantiate VMs on the same data domain where the
these two seem to prevent cloudforms from being used efficiently in such
are there workarounds I'm not aware of? work in progress? worth an RFE?
Giulio Fidente, gpg key id 08D733BA
This is the text for the "SysAdmin" Persona (Sam), presented a
few minutes ago.
Trying to think of how to specify needs/wants, but nothing's
coming to mind. Any suggestions?
Sysadmin Persona: Sam
* Sam is a 35 year old male from Hanoi, Vietnam, who migrated
to east coast US about 10 years ago.
* He is the sole breadwinner in his family, happily married to
his wife of 8 years, and has a 3 year old child.
* Sam works as the Aeolus SysAdmin for "T-Bone", a large
monopoly style Telco in his area.
* He previously worked on Enterprise Class Solaris Systems for
many years, and moved to working with Linux (mostly RHEL), a
few years ago.
* "T-Bone" has dedicated NAS, SAN and network teams, with a 2
week+ (min) turn around time for requests.
* Aeolus runs on a single server in this environment, a dual
cpu (2 x quad core) HP DL380 with 24 GB RAM, and 2TB of
local RAID 0 SATA storage.
* Aeolus is hooked up to a small vSphere 5.x cluster. vCenter
is running on a server identical to the Aeolus one, and 4 x
hosts (also HP DL 380's) with 96GB RAM and 4TB of shared
NetApp storage (a fast NAS).
* Sam's team also use Aeolus with EC2, for giving their
partners access to selected builds of their applications,
for testing and qualification (not prod).
* The Aeolus environment is used by the developers in a part
of T-Bone's business and government division. About 40
* The standard development platform the developers write for
is RHEL 6.x, JBoss (various releases), and Oracle 10g/11g.
* NetBackup is the enterprise backup solution, so Sam wrote
custom shell scripts to do backups of Aeolus nightly.
They don't back up generated Aeolus images, but do backup
up everything else (Aeolus state, XML templates, etc).
* Sam has a ToDo list item to investigate RHEV, and has been
toying around with oVirt on a spare desktop box. Not in
depth with it yet.
Needs / wants
Aeolus Community Manager
We had a good discussion yesterday here in Brno on the mission of
Aeolus. Four teams went off and came up with four different statements,
which we then jammed together to produce the result below. Let us know
what you think -- suggestions are welcome, but please be constructive.
"Aeolus mission: to provide superior tools and workflows for flexible
construction, management, and monitoring of multi-instance systems
== Hugh Brock, hbrock(a)redhat.com ==
== Engineering Manager, Cloud BU ==
== Aeolus Project: Manage virtual infrastructure across clouds. ==
== http://aeolusproject.org ==
"I know that you believe you understand what you think I said, but I’m
not sure you realize that what you heard is not what I meant."
Hi aeolus community,
We've set some dates for the upcoming sprint to work with the week of planning and discussion in Brno next week.
Please take a look and review important dates here 
These are the dates of the most immediate interest to developers:
Official start of sprint
All patches to list or pull requests made
Thanks for everything you do!
Looking forward to a productive sprint,
-------- Original Message --------
Subject: Announcing Deltacloud Community Call
Date: Fri, 02 Nov 2012 14:50:02 +0200
From: marios(a)redhat.com <mandreou(a)redhat.com>
we think it might be useful to have a weekly open/community call to
discuss all things Deltacloud. The proposal is for 1 hour calls, the
format of which is up for discussion. The idea right now is to have 2
'topics' per call; topics can be:
1. A presentation on a particular aspect - 'How do exceptions work in
2. A question about something you're working on - 'JIRA ticket
DTACLOUD_555 - discussion and possible approaches'
3. An 'open' Q/A session with the devs.
4. Anything else you want/think might be useful...
If on a given week there are no topics then we can just have an open
discussion on "what i'm working on and how I've implemented it" for
early feedback for example.
So, as we are all all over the place - it may be tricky to find a
suitable time. An entirely unscientific poll of a few people has
indicated that the 0800 Pacific time may be a good slot... making it
1600 in Europe and 1100 in NYC. It also looks like Thurdsay/Friday would
be a good day due to conflicts earlier in the week. Please indicate your
preference on the poll @ http://doodle.com/c2ht9zf4e7gngrhu (be sure
to change the time zone to whatever is local to you). If you can't make
*any* of those times then just vote no to everything and reply to this
mail with suggestions.
A preliminary etherpad for the first call is @
Comments and suggestions pretty please?