[ACTION REQUIRED] Any ovirt-engine master/3.6/3.6.1 patches must be rebased to include automation changes
by David Caro
--OGW1Z2JKiS9bXo17
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
In order to make more generic and to allow migrating the artifactory server
that currently is providing us with the artifacts to build and test
ovirt-engine, the port of the url we were using has been changed:
master -> https://gerrit.ovirt.org/#/c/49275/
3.6 -> https://gerrit.ovirt.org/#/c/49276/
3.6.1 -> https://gerrit.ovirt.org/#/c/49277/
The migration of the artifactory server will be done later next week, so if=
you
don't want your ci tests to start failing at that point, please rebase any
patches that you might have for those branches on top of those commits (they
are merged already).
This also allown to avoid having to change those scripts anymore for future
migrations
Thanks!
--=20
David Caro
Red Hat S.L.
Continuous Integration Engineer - EMEA ENG Virtualization R&D
Tel.: +420 532 294 605
Email: dcaro(a)redhat.com
IRC: dcaro|dcaroest@{freenode|oftc|redhat}
Web: www.redhat.com
RHT Global #: 82-62605
--OGW1Z2JKiS9bXo17
Content-Type: application/pgp-signature; name="signature.asc"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAEBAgAGBQJWWCc9AAoJEEBxx+HSYmnDqxwH/iRZF2RFiR0fc3MloC3fRdfg
40UMgUmEy50kaEn0Jb7SlwrtvqiA4zA4tCH1qXz5UXm6+gEVgBUVQYPzOPdYlYA/
JvIM5ytmOr+5mBrSlmTYjSbdBd2xnK3uz1WirNjrlfRUIavuGKXI+UdeHaSKyb0F
EmjhhLNroEpVxV4wuEEU8ZyM7ia9jqfuVsE0763N/eh01MN7E+GvTBhmw4XmFKJU
nObLeK3sxS8ye+65ujP4xw7CyO6Q28U7q2Kk/7qeKFs4PYIbhKwU+AT4/d2sLjXb
Hy1koCZpO04WdcQWqCplX9P14GCMZONels0cSj52KejZ6tXwvbegdLyXk86TiDw=
=Uvba
-----END PGP SIGNATURE-----
--OGW1Z2JKiS9bXo17--
8 years, 5 months
[oVirt 3.6 Localization Question #40] "This field must contain an {0} number"
by Yuko Katabami
Hello once again,
I would like to ask you another question.
File: UIMessages
Resource ID: thisFieldMustContainTypeNumberInvalidReason
String: This field must contain an {0} number
Question: I can see from the source comment 0=type but could you please
give me an example of what will replace {0}?
Kind regards,
Yuko
8 years, 5 months
[oVirt 3.6 Localization Question #41] "This field can be empty or contain a {0}"
by Yuko Katabami
Hello all,
I am posting another question raised by a German translator.
Could anyone help us?
File: UIMessages
Resource ID: emptyOrValidKeyValueFormatMessage
String: This field can be empty or contain a {0}
Question: Could you please let me know what will replace {0}? As there are
multiple articles in Latin languages such as German and French, we would
need to know the gender of the noun in order to select the correct article
for this accusative object.
Kind regards,
Yuko
8 years, 5 months
[oVirt 3.6 Localization Question #38] "< key >=< value > separated with space"
by Yuko Katabami
Hello again,
I have another question I would like to ask for your help:
File: uicompat.UIMessages
Resource ID: keyValueFormat
String: < key >=< value > separated with space
Question: Is this asking user to provide a space-separated list of multiple
"< key >=< value >" values?
Kind regards,
Yuko
8 years, 5 months
Push notifications in 4.0 backend
by Martin Betak
Hi All,
I would like to take this opportunity to start a discussion
about the possibility of implementing a user facing change notifications.
The benefit of this would be to remove the need for periodic polling
from frontends and other services that consume our REST API.
Also implmenting a common infrastructure on the backend for event
notifications (e.g. CDI events) would further reduce the internal
need for polling the DB by the backend itself, maybe even reducing
the need to use DB for some things and just keep them in memory and
updated by CDI event observers.
There are many solutions how to provide the user-facing part of the notifications:
Doctor Rest, MQTT, websocket, server-sent events, ... . Ideally these notifications
should be consumable both by web browser (webadmin/userportal) but also by
other services (such as ManageIQ), or other REST clients such as moVirt android client.
But regardless of the chosen user-facing transport, I believe a common infrastructure
can be implemented on the BLL layer with the usage of CDI events fired from commands.
I see 2 major sources of changes in the engine (please correct me if this is wrong):
1) CRUD & management commands
2) Vms/Hosts monitoring
the changes originating from 2) are AFAIK very localized and not so numerous so a manual
firing of appropriate events for VMs and Hosts could be added here.
The 1) case is more extensive in terms of required code changes. While a manual solution
would still be feasible, I believe there is place for a more automated/declarative way.
One solution for 1) that comes to my mind are simple command-level annotations covering the
Created, Updated, Removed (C, U, and D from CRUD) cases. The goal here is to fire the
appropriate CDI events when an entity is created/updated/deleted. Since commands usually
contain getters for entities they work with (such as getVm(), getVds(), getStorageDomain() ...)
It should be sufficient for the most common simple cases (of course this will not cover
everything) to use annotation @Creates, @Updates, @Removes on the commands classes, where
parameters of the annotation should specify the getter method that returns the affected entity
(VM/VDS/StorageDomain...). This could be specified by the entity class token or method name
(depending on the level of "magic" one prefers :-) and the CommandBase infrastructure would
then collect those annotations and upon successful completion of the command fire the
appropriate events.
Example #1:
@Updates('getVm') // or @Updates(VM.class)?
public class UpdateVmCommand<...> extends VmManagementComandBase ...
Note that since Java 8 we have repeatable annotations so we can have more complex commands
that affect more entities.
Example #2:
@Updates(Vm.class)
@Updates(VmTemplate.class)
// possibly also some @Creates and @Removes annotations or their combination
public class ContrivedExampleCommand extends SomeCommandBase
the infrastructure would then look upon successful completion of the command on the getVm()
and getVmTemplate() methods, invoke them, determine the resulting types of entities VM and VmTemplate
and since the annotations used were @Updates fire CDI event equivalent to
@Inject
@Updated // our custom CDI qualifier
Event<VM> vmChangedEvent;
and anologously for VmTemplate.
But regardless of the exact implementation of the CDI event firing: whether manual, the above
proposal, or some crazy usage of AspectJ - the interface for the rest of the code should always
be the like this:
public void onVmChanged(@Observes @Updated VM vm) {
// ....
}
On top of this, I believe, we can build the user-facing part of push notifications and also
improve our existing backend code.
Thank you for reading this long email and I welcome any comments or counter-proposals you
might have on this topic :-)
Best regards,
Martin
8 years, 5 months
[ANN] oVirt 3.6.1 First Release Candidate is now available for testing
by Sandro Bonazzola
The oVirt Project is pleased to announce the availability
of the First Release Candidate of oVirt 3.6.1 for testing, as of November
25th, 2015.
This release is available now for Fedora 22,
Red Hat Enterprise Linux 6.7, CentOS Linux 6.7 (or similar) and
Red Hat Enterprise Linux >= 7.1, CentOS Linux >= 7.1 (or similar).
This release supports Hypervisor Hosts running
Red Hat Enterprise Linux >= 7.1, CentOS Linux >= 7.1 (or similar) and
Fedora 22.
Highly experimental support for Debian 8.1 Jessie has been added too.
This release of oVirt 3.6.1 includes numerous bug fixes.
See the release notes [1] for an initial list of the new features and bugs
fixed.
Please refer to release notes [1] for Installation / Upgrade instructions.
A new oVirt Live ISO is also available[2].
Please note that mirrors[3] may need usually one day before being
synchronized.
Please refer to the release notes for known issues in this release.
[1] http://www.ovirt.org/OVirt_3.6_Release_Notes
[2] http://resources.ovirt.org/pub/ovirt-3.6-pre/iso/
[3] http://www.ovirt.org/Repository_mirrors#Current_mirrors
--
Sandro Bonazzola
Better technology. Faster innovation. Powered by community collaboration.
See how it works at redhat.com
8 years, 5 months
oVirt Node roadmap
by Fabian Deutsch
Hey,
in the last few months the Node team spent a lot of efforts in
stabilizing Node by closing tons of bugs, rebasing Node onto CentOS
7.2, and in addition adding features like Hosted-Engine - to get Node
in shape for the recently released oVirt 3.6.0.
But we were also seeing how Node is showing it's age. It becomes more
and more challenging to solve bugs and implement features in Node's
current architecture.
To address these problems, and let Node adjust better to new changes,
a few months ago we started to look at how we can change Node, to make
it easier to develop, test, and use.
This comparison [1] shows a brief summary of our investigations. We
especially looked at Atomic and containers [2].
At the bottom line both of them provided us an architecture which
would help us to achieve something like 70% of what we need. But
during early trials we quickly ran into issues which we experience in
similar ways with today's Node.
Finally we settled an approach - the idea was around since right from
the beginning - which aligns very well with existing technologies
which we already use in the oVirt and Fedora/CentOS scope.
The new Node will be using anaconda for installation, LVM for upgrades
and rollbacks, and Cockpit [3] for administration. The future design
is taking care that packages don't need to take special care to work
on Node - which was a major obstacle in the past. Node will rather
behave (mostly) like a regular host - but with the advantages of an
easy & ready to use image, image based delivery and a robust rollback.
The current design principles and links to some additional resources
are in the wiki [4].
Stay tuned, we are just getting started.
On behalf of the Node Team
fabian
--
[1] http://www.ovirt.org/Node/Specs#Comparison:_Possible_Implementations
[2] http://www.projectatomic.io/ ~ http://docker.com/
[3] http://cockpit-project.org/
[4] http://www.ovirt.org/Node/4.0
8 years, 5 months
