This is essentially a revert of
73695d0ea490c4c7a1158957dd5a85586cfa0933. Since stackrc is
responsible for sourcing localrc, the required functions
are necessary (like for enable_service).
fixes bug 1186488
Change-Id: Iad90f802e77b94416821008c294c1a2ede8a4729
Set [database]connection option to mongodb rather than sqlite as mongodb
is the only fully supported db.
Change-Id: I2b9fbc65de4476593d456a782dbc7e6540a75141
Fixes: bug #1186240
The Nova v3 API will be disabled by default after 31164 goes through.
This change explicitly enables the v3 API in the nova config file
Change-Id: I8b4c64b4718fc96d7a928c6fa3a0e8716c7edba1
Simple check to see if EPEL repo is enabled for RHEL6 and, if not,
automated install of the repo RPM.
Also adds an additional sanity check which checking for pip. In the
prior case of EPEL being disabled, a RHEL6 host may not have access to
python-pip. Although this shouldn't happen, its a good sanity check.
Change-Id: I4e8a4dda9475b75d071534d4eef469198502a048
nodejs is required for Horizon (ships a copy of lessc internally). I
couldn't find nodejs in the rpm list, despite it being installed in
lib/horizon. This adds to the rpm list as NOPRIME and retains the old
install point.
RHEL6 has nodejs in EPEL now, so we remove the old hack that installed
an upstream version and add it to the install mentioned previously
Change-Id: Ib3ee2f7d77e22666c9055c8528288c01b46a95d3
rejoin-stack.sh would not specify screen session name which will lead
unstack.sh unable to kill the screen session started by rejoin-stack.sh
after hypervisor rebooted. So specify the session name if
rejoin-stack.sh need to start a new sreen session.
fixes bug 1182806
Change-Id: I68635887aa82fa63f3667e0f090ea2836eec7b68
Currently lib/heat says:
To enable, add the following to localrc
ENABLED_SERVICES+=,heat,h-api-cfn,h-api-cw,h-eng
Once the stack is up, `heat list` fails because the API server
isn't running.
This commit adds h-api to that list.
Also make sure h-api is killed in stop_heat().
Change-Id: I2e818bb343680b3778f9277c23c766f784d28887
* It's not possible to prevent python-nose (which is not compatible
with Tempest) from being installed since so many os packages
dependent on it.
* This patch installs python-nose1.1 from EPEL and adds a
/usr/local/bin symlink for nosetests1.1 so that invocations for
Tempest will work as expected.
Change-Id: I756d1ca5eda127dcdd2efb047c8e14dd344f710d
This patch makes it easier to support the all-in-one installation of
XenServer with Quantum:
- separate agents for domU and dom0 openvswitches. The domU agent's name
is "q-domua", and the dom0 agent is called "q-agt"
- create a physnet1 physical network to connect those openvswitches
Note: This patch depends on the quantum patch:
https://review.openstack.org/15022
For a diagram, see:
http://goo.gl/BuAdg
Change-Id: Iff851438553528b9051edd380ecbb2c0415cbf78
Switching to the redis matchmaker.
The localhost matchmaker does not presently work.
Also, the localhost matchmaker could not work for
multi-host setups.
Change-Id: I81a26b9af78328e360a18b87371c619e194365d2
By default, the zeromq driver looks for
/var/run/openstack which needs to be created and
given the right permissions.
It is easier and just as safe in the case of devstack
to just use mktemp to establish a temporary working
directory.
Change-Id: I4cec33e49d2b042a244420fb40d83d476e4971cd
The Xenserver/XCP part of devstack was configuring the hypervisor's
connectivity to the outer world, by adding VLANs, and physical
interfaces to the virtual networks. It added a lot of complexity, and
made it hard to get started with XenServer. This patch removes that
extra complexity, so it is left as an exercise for the user.
Related to blueprint blueprint xenapi-devstack-cleanup
Change-Id: If3367335c3da8621d0afe1f6cae77511fbdbb3e2
realized that apparently I'd never done that, assuming it was
auto generated like other projects. Oops.
Change-Id: Ib3374e1eed54fb723da8afc006f83d42432307aa
Use .mailmap file to coalesce commits
by the same person using multiple names and/or email addresses
Change-Id: I00c060d62a4671ca6ce0a5e04ac3c4f144ed671e
fixes bug 1179820
we must set firewall driver if we want to use quantum security group because
quantum will disable security group if we do not change the default firewall
driver. Currently devstack will not change the default firewall driver
if we just running quantum server on a node which will cause nova unable to
security group information.
Change-Id: Ie274325decbf252630a237ed3d6ee3136eb259fe
When devstack creates stack-screenrc, it includes unescaped "
characters that cause failures starting nova-compute. This fix
changes the " to ' so there isn't a conflict.
Fixes bug 1183114
Change-Id: I7830879d56f1ac20950aace46dd3b72d209986ce
When Quantum auth_token configuration were moved to quantum.conf,
the temporary logic to disable auth_token in api-paste.ini was added
to avoid gating test failures of patches based on the version with
api-paste.ini auth_token configurations.
This temporary logic is no longer needed.
Change-Id: I539301a338d5ada01c83af5e9cf203849a67c6c1
The other services, like nova, glance and cinder, use SERVICE_HOST
instead of HOST_IP to configure their server listening addresses.
This is more friendly to multi-node quantum deployment.
Change-Id: Ibee22ef59d532e8f7004d8e4582015d303d15404
* RHEL6 does not include Open vSwitch in the default repos, but it
is available via the RDO repo. This patch automatically
configures the RDO repo for RHEL6.
* The openvswitch package is now listed as an rpm dependency to
ensure it can be cached by a pre-built CI environment.
Change-Id: I8f93f53039cca4ff29bcb91720be7bb047b3ed8a
If a virtual machine has got a managed save image, it is not
possible to delete it using a plain 'virsh undefine' command.
While Nova doesn't use 'managed save', the libvirt-guests
init script or systemd service may have created one if the
user rebooted their host. Thus we devstack should pass the
--managed-save flag to virsh to ensure the VM's removal.
Change-Id: Id9b072a6dceeb4e179d2b6a25bbdfb559c299e95
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
As discovered in tempest, when you create and delete volumes
quickly, and have a small volume storage pool, you can get cinder
scheduler to think you are "out of space" when you are not.
This is because cinder scheduler updates free space on it's
periodic job, which defaults to 60 seconds. We need control over
that value for the devstack gate otherwise we regularly get
overruns if we run too many volume tests quickly.
Work around for bug 1180976
Expect that this will get removed later if cinder gets a more
efficient way to update the scheduler for freed resources.
Also, don't be completely stupid about setting defaults....
Change-Id: I20e52e66fcc94b224476cdd14c88bd6981b4e617
* Execution of the setenforce command when selinux was not enabled
was resulting in a non-zero exit status, which halted devstack.
* Addresses bug 1175444
Change-Id: I7f6492dea7c52d153e755cda826d6fabd53f3771