| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This patch moves all of the os-apply-config (tripleo-image-elements)
specific templates into a common directory. This matches what
we do for puppet and should help new users more easily
understand the project layout.
Change-Id: I7dce2a770d56795f3ea22c8a464595c4a0c60832
|
|
Adds hook to enable additional "AllNodes" config to be performed prior
to applying puppet - this is useful when you need to build
configuration data which requires knowledge of all nodes in a cluster,
or of the entire deployment.
As an example, there is a sample config template which collects the
hostname and mac addresses for all nodes in the deployment then writes
the data to all Controller nodes. Something similar to this may be
required to enable creation of the nexus_config in
https://review.openstack.org/#/c/198754/
There's also another, simpler, example which shows how you could share
the output of an OS::Heat::RandomString between nodes.
Change-Id: I8342a238f50142d8c7426f2b96f4ef1635775509
|
|
|
|
When using network isolation you might want to selective
move one of the services back to the default ctlplane network
by simply using the ServiceNetMap parameter. This patch
adds ctlplane to the output parameters for both
the net_ip_map and net_ip_list_map nested stacks so that
this is possible.
As part of this patch we also split out the NetIpSubnetMap
into its own unique nested stack so that the Heat input
parameters for this stack are more clearly named.
Change-Id: Iaa2dcaebeac896404e87ec0c635688b2a59a9e0f
|
|
This patches wires in a new "all nodes" validation resource
that can be used to add validations that occur early on
during the deployment process. This occurs after the nodes
have been brought online and the initial networks
have been configured but before any "post" (puppet, etc.)
sort of configuration has been executed.
A initial validation script has been added to ping test network IPs
on each network. When using network isolation this will ensure
network connectivity (vlans, etc) are working on each
node and if not the heat stack will fail early, allowing
time to fix the network connections and retry the
stack creation via an update.
Change-Id: I63cf95b27e8ad2aed48718cf84df5f324780e597
Co-Authored-By: Ian Main <imain@redhat.com>
Co-Authored-By: Ryan Hallisey <rhallise@redhat.com>
|
|
The redis_vip should come from a Neutron Port as its cidr depends
on the Neutron Network configuration. This change adds 2 new files
and modifies 1 in the network/ports directory:
- noop.yaml - Passes through the ctlplane Controller IP (modified)
- ctlplane_vip.yaml - Creates a new VIP on the control plane
- vip.yaml - Creates a VIP on the named network (for isolated nets)
Also, changes to overcloud-without-mergepy.yaml create the
Redis Virtual IP. The standard resource registry was modified to
use noop.yaml for the new Redis VIP. The Puppet resource registry
was modified to use ctlplane_vip.yaml by default, but can be made
to use vip.yaml when network isolation is used by using an
environment file. vip.yaml will place the VIP according to the
ServiceNetMap, which can also be overridden.
We use this new VIP port definition to assign a VIP to Redis,
but follow-up patches will assign VIPs to the rest of the
services in a similar fashion.
Co-Authored-By: Dan Sneddon <dsneddon@redhat.com>
Change-Id: I2cb44ea7a057c4064d0e1999702623618ee3390c
|
|
This change adds config and deployment resources to trigger package
updates on nodes. The deployments are triggered by doing a stack-update
and setting one of the parameters to a unique value.
The intent is that rolling update will be controlled by setting
breakpoints on all of the UpdateDeployment resources inside the
role resource groups.
Change-Id: I56bbf944ecd6cbdbf116021b8a53f9f9111c134f
|
|
This patch adds VIPs for the internal_api, storage,
and storage management networks.
For puppet these are persisted into a local vip-config
hieradata file which is then used by puppet-tripleo's
loadbalancer module to apply per-service VIP settings.
Change-Id: I909c3bdc9d17a8e15351f4797287769e3f76c849
|
|
This patch adds a new NetIpListMap abstraction which we can use
to make the all-nodes-config IP list network assignments
configurable. Ip address lists for all overcloud services
which require IPs were added to all-nodes-config so
that puppet manifests can be directly supplied the
correct network list for each service.
Change-Id: I209f2b4f97a4bb78648c54813dad8615770bcf1a
|
|
This patch adds a resource which constructs a Json output
parameter called net_ip_map which will allow us to easily
extract arbitrary IP addresses for each network using the
get_attr function in heat.
The goal is to use this data construct in each role
template to obtain the correct IP address on each
network.
Change-Id: I1a8c382651f8096f606ad38f78bbd76314fbae5f
|
|
This patch updates the cinder block storage roles so that
they can optionally make use of isolated network
ports on the storage, storage management, and internal_api
networks.
-Multiple networks are created based upon settings in the heat
resource registry. These nets will either use the noop network (the
control plane pass-thru default) or create a custom Neutron port on
each of the configured networks.
-The ipaddress/subnet of each network is passed passed into the
NetworkConfig resource which drives os-net-config. This allows the
deployer to define a custom network template for static IPs, etc
on each of the networks.
-The ipaddress is exposed as an output parameter. By exposing
the individual addresses as outputs we allow Heat to construct
collections of ports for various services.
Change-Id: I4e18cd4763455f815a8f8b82c93a598c99cc3842
|
|
This patch updates the swift roles so that
they can optionally make use of isolated network
ports on the storage, storage management, and internal API
networks.
-Multiple networks are created based upon settings in the heat
resource registry. These nets will either use the noop network (the
control plane pass-thru default) or create a custom Neutron port on
each of the configured networks.
-The ipaddress/subnet of each network is passed passed into the
NetworkConfig resource which drives os-net-config. This allows the
deployer to define a custom network template for static IPs, etc
on each of the networks.
-The ipaddress is exposed as an output parameter. By exposing
the individual addresses as outputs we allow Heat to construct
collections of ports for various services.
Change-Id: I9984404331705f6ce569fb54a38b2838a8142faa
|
|
This patch updates the ceph roles so that
they can optionally make use of isolated network
ports on the storage and storage management networks.
-Multiple networks are created based upon settings in the heat
resource registry. These nets will either use the noop network (the
control plane pass-thru default) or create a custom Neutron port on
each of the configured networks.
-The ipaddress/subnet of each network is passed passed into the
NetworkConfig resource which drives os-net-config. This allows the
deployer to define a custom network template for static IPs, etc
on each of the networks.
-The ipaddress is exposed as an output parameter. By exposing
the individual addresses as outputs we allow Heat to construct
collections of ports for various services.
Change-Id: I35cb8e7812202f8a7bc0379067bf33d483cd2aec
|
|
This patch updates the compute roles so that
they can optionally make use of isolated network
ports on the tenant, storage, and internal_api networks.
-Multiple networks are created based upon settings in the heat
resource registry. These nets will either use the noop network (the
control plane pass-thru default) or create a custom Neutron port on
each of the configured networks.
-The ipaddress/subnet of each network is passed passed into the
NetworkConfig resource which drives os-net-config. This allows the
deployer to define a custom network template for static IPs, etc
on each of the networks.
-The ipaddress is exposed as an output parameter. By exposing
the individual addresses as outputs we allow Heat to construct
collections of ports for various services.
Change-Id: Ib07b4b7256ede7fb47ecc4eb5abe64b9144b9aa1
|
|
This patch updates the controller roles so that
they can optionally make use of isolated network
ports on each of 5 available overcloud networks.
-Multiple networks are created based upon settings in the heat
resource registry. These nets will either use the noop network (the
control plane pass-thru default) or create a custom Neutron port on
each of the configured networks.
-The ipaddress/subnet of each network is passed passed into the
NetworkConfig resource which drives os-net-config. This allows the
deployer to define a custom network template for static IPs, etc
on each of the networks.
-The ipaddress is exposed as an output parameter. By exposing
the individual addresses as outputs we allow Heat to construct
collections of ports for various services.
Change-Id: I9bbd6c8f5b9697ab605bcdb5f84280bed74a8d66
|
|
This patch enables uses to selectively enable the creation
of split out networks for the overcloud traffic. These
networks will be created on the undercloud's neutron
instance.
By default a noop network is used so that no extra networks
are created. This allows our default to continue being
all traffic on the control plane.
Change-Id: Ied49d9458c2d94e9d8e7d760d5b2d971c7c7ed2d
|
|
|
|
This change allows a different network config for each family of hosts. For
instance, the controller may have a different network configuration than a
block storage node. This change adds a declaration for each family in the
overcloud-resource-registry.yaml & overcloud-resource-registry-puppet.yaml.
Change-Id: I083df7ebbb535f97d8ddec2ac0e06281c55986cd
|
|
Adds optional hooks which can run operator defined additional config on
nodes after the application deployment has completed.
Change-Id: I3f99e648efad82ce2cd51e2d5168c716f0cee8fe
|
|
Currently all the OS::Nova::Server resource created don't pass any
user-data. It's possible to pass user-data as well as using heat
SoftwareConfig/SoftwareDeployment resources, and this can be useful
when you have simple "first boot" tasks which are possible either via
cloud-init, or via simple run-once scripts.
This enables passing such data by implementing a new provider resource
OS::TripleO::NodeUserData, which defaults to passing an empty mime
archive (thus it's a no-op). An example of non no-op usage is also
provided.
Change-Id: Id0caba69768630e3a10439ba1fc2547a609c0cfe
|
|
This is a first implementation of Ceph support in TripleO with Puppet:
* Install ceph-mon on controller node
* Install ceph-osd on cephstorage node
Co-Authored-By: Giulio Fidente <gfidente@redhat.com>
Change-Id: I48488cbe950047fae5e746e458106d6edb9a6183
|
|
This patch adds a new BlockStoreNodesPostDeployment resource
which can be used along with the environment file to
specify a nested stack which is guaranteed to execute
after all the BlockStore config deployments have executed.
This is really useful for Puppet in that Heat actually
controls where puppet executes in the deployment
process and we want to ensure puppet runs after
all hiera configuration data has be deployed to
the nodes. With the previous approach some of the
data would be there, but allNodes data would not be
guaranteed to be there in time.
As os-apply-config (tripleo-image-elements) have their
ordering controlled within the elements themselves an empty stubbed
in nested stack has been added so that we don't break that
implementation.
Change-Id: I29b3574e341eecd53b2867788f415bff153cfa9f
|
|
This patch adds a new ObjectStoreNodesPostDeployment resource
which can be used along with the environment file to
specify a nested stack which is guaranteed to execute
after all the ObjectStore config deployments have executed.
This is really useful for Puppet in that Heat actually
controls where puppet executes in the deployment
process and we want to ensure puppet runs after
all hiera configuration data has be deployed to
the nodes. With the previous approach some of the
data would be there, but allNodes data would not be
guaranteed to be there in time.
As os-apply-config (tripleo-image-elements) have their
ordering controlled within the elements themselves an empty stubbed
in nested stack has been added so that we don't break that
implementation.
Change-Id: I778b87a17d5e6824233fdf9957c76549c36b3f78
|
|
This patch adds a new ComputeNodesPostDeployment resource
which can be used along with the environment file to
specify a nested stack which is guaranteed to execute
after all the Compute config deployments have executed.
This is really useful for Puppet in that Heat actually
controls where puppet executes in the deployment
process and we want to ensure puppet runs after
all hiera configuration data has be deployed to
the nodes. With the previous approach some of the
data would be there, but allNodes data would not be
guaranteed to be there in time.
As os-apply-config (tripleo-image-elements) have their
ordering controlled within the elements themselves an empty stubbed
in nested stack has been added so that we don't break that
implementation.
Change-Id: I80bccd692e45393f8250607073d1fe7beb0d7396
|
|
This patch splits out the BootstrapNode config
such that alternate implementation (puppet for example)
can implement their own SoftwareConfig's via a nested stack.
This is controlled by the standard overcloud heat environment.
For os-apply-config deployments the implementation should work the
same as before.
For puppet deployments the implementation uses hiera metadata
to configure bootstrap_nodeid.
Change-Id: I691a9d7c474866038a5d47beab295899b5479d03
|
|
This patch splits out the allNodesConfig config
such that alternate implementation (puppet for example)
can implement their own SoftwareConfig's via a nested stack.
This is controlled by the standard overcloud heat environment.
For os-apply-config deployments the implementation should work the
same as before.
For puppet deployments the implementation uses hiera metadata
to configure rabbit_nodes. The puppet deployment doesn't support
hosts, or freeform sysctl metadata yet so those are the same
for now as well.
Change-Id: I34ae30b1f37aca8b39586f7e350511462d66f694
|
|
This patch splits out the SwiftDevicesAndProxy config
such that alternate implementation (puppet for example)
can implement their own SoftwareConfig's via a nested stack.
This is controlled by the standard overcloud heat environment.
For os-apply-config deployments the implementation should work the
same as before.
For puppet deployments the implementation uses hiera metadata
to configure swift devices.
Partial-bug: 1418805
Change-Id: Ibf6038460f36279ad51a04947589d4a03a553f66
|
|
This patch adds a new ControllerNodesPostDeployment resource
which can be used along with the environment file to
specify a nested stack which is guaranteed to execute
after all the Controller config (HA, or other) have
executed.
This is really useful for Puppet in that Heat actually
controls where puppet executes in the deployment
process and we want to ensure puppet runs after
all hiera configuration data has be deployed to
the nodes. With the previous approach some of the
data would be there, but most of the HA data which
actually gets composed outside of the controller-puppet.yaml
nested stack would not be guaranteed to be there in time.
As os-apply-config (tripleo-image-elements) have their
ordering controlled within the elements themselves an empty stubbed
in nested stack has been added so that we don't break that
implementation.
Partial-bug: 1418805
Change-Id: Icd6b2c9c1f9b057c28649ee3bdce0039f3fd8422
|
|
The new ceph-source.yaml file provides the config settings needed
by the elements which configure Ceph on controllers (monitors) and
storage nodes (OSDs) as well as the Cinder backend which uses it.
There is also a without-mergepy copy named ceph-storage.yaml
Change-Id: I954861536c41b2a7e6cbd86a0f0b55004eed4c70
|
|
In I250dc1a8c02626cf7d1a5d2ce92706504ec0c7de we split
out just the Controller software config in an effort
to provide hooks for alternate implementations (puppet).
This sort of worked but caused quirky ordering issues
with signal handling. It also causes problems for Tuskar
which would prefer to think of these nested stacks and
not have us split out just the software configs like this.
This patch moves all the compute related stuff for
our two implementations:
compute.yaml: is used by os-apply-config (uses the
tripleo-image-elements)
compute-puppet.yaml: uses stackforge puppet-* modules for
configuration
By duplicating the entire compute in this manner we make
it much easier to create dependencies and implement proper
signal handling. The only (temporary) downside is the duplication
of parameters most of which will eventually go away when we move
using the global parameters via Heat environment files instead.
Change-Id: I49175d1843520abc80fefe9528442e5dda151f5d
|
|
In I228216a0b55ff2d384b281d9ad2a61b93d58dab9 we split
out just the Controller software config in an effort
to provide hooks for alternate implementations (puppet).
This sort of worked but caused quirky ordering issues
with signal handling. It also causes problems for Tuskar
which would prefer to think of these nested stacks and
not have us split out just the software configs like this.
This patch moves all the controller related stuff for
our two implementations:
controller.yaml: is used by os-apply-config (uses the
tripleo-image-elements)
controller-puppet.yaml: uses stackforge puppet-* modules for
configuration
By duplicating the entire controller in this manner we make
it much easier to create dependencies and implement proper
signal handling. The only (temporary) downside is the duplication
of parameters most of which will eventually go away when we move towards
using the global parameters via Heat environment files instead.
Change-Id: Iaf3c889d7c8815f862308cd8e15ce1010059f5c6
|
|
This patch provides an alternate implementation of
the OS::TripleO::Controller::SoftwareConfig which uses Puppet
to drive the configuration. Using this it is possible
to create a fully functional overcloud controller instance
which has the controller node configured via Puppet
stackforge modules. Initially this includes only the
following services:
MySQL
RabbitMQ
Keepalived/HAProxy (HA is not yet fully supported however)
Nova
Neutron
Keystone
Glance (file backend)
Cinder
Using these services it is possible to run devtest_overcloud.sh
to completion. The idea is that we can quickly add more
services once we have CI in place.
In order to test this you'll want to build your images
with these elements:
os-net-config
heat-config-puppet
puppet-modules
hiera
None of the OpenStack specific TripleO elements
should be used with this approach (the nova/neutron
elements were NOT used to build the controller image).
Also, rather than use neutron-openvswitch-agent to configure
low level networking it is recommended that os-net-config
by configured directly via heat modeling rather than
parameter passing to init-neutron-ovs. This allows us to
configure the physical network while avoiding the coupling to
the neutron-openvswitch-element that our standard
parameter driven networking currently uses. (We still need
to move init-neutron-ovs so that it isn't coupled and/or deprecate
its use entirely because the heat drive stuff is more flexible.)
Packages may optionally be pre-installed via DIB using the
-p option (-p openstack-neutron,openstack-nova) etc.
Change-Id: If8462e4eacb08eced61a8b03fd7c3c4257e0b5b8
|
|
This is a step towards supporting pluggable software configurations
in the heat templates. By moving controller-config out of controller.yaml
we make it possible to define alternate implementations by
changing the OS::TripleO::ControllerConfig value in the
overcloud-resource-registry.yaml heat environment file.
Change-Id: I228216a0b55ff2d384b281d9ad2a61b93d58dab9
|
|
This example extends the compute software configuration
so that heat metadata is used to model the os-net-config
YAML (ultimately JSON) directly. The existing
os-net-config element already supports this format.
Configuring the physical network layer in this manner
would supplant the ever growing list of Heat parameters
that we have and is something that could be automatically
generated via tuskar.
The default is to use net-config-noop.yaml which
will pass no config metadata into the os-net-config
element which will essentially disable it in favor
of using parameters w/ init-neutron-ovs.
Change-Id: I30f325b1751caaef5624537e63ee27c2e418d5c8
|
|
This is a step towards supporting pluggable software configurations
in the heat templates. By moving compute-config out of compute.yaml
we make it possible to define alternate implementations by
changing the OS::TripleO::Compute::SoftwareConfig value in the
overcloud-resource-registry.yaml heat environment file.
Co-Authored-By: Steve Hardy <shardy@redhat.com>
Change-Id: I250dc1a8c02626cf7d1a5d2ce92706504ec0c7de
|
|
This patch extends the previous 'Don't use merge.py for overcloud'
commit with the cinder-storage.yaml and swift-storage.yaml templates.
Requirements for this to deploy:
1. Block and object storage images have to be built
(overcloud-cinder-volume and overcloud-swift-storage)
2. The images have to be loaded by devtest_overcloud.sh
OVERCLOUD_CINDER_ID=$(load-image -d $TRIPLEO_ROOT/overcloud-cinder-volume.qcow2)
OVERCLOUD_SWIFT_ID=$(load-image -d $TRIPLEO_ROOT/overcloud-swift-storage.qcow2)
Change-Id: I45f9d9f051970a83e26c0fd924d7c98276958113
|
|
This provides three templates: overcloud-without-mergepy.yaml,
compute.yaml and controller.yaml. These can be used in combination with
overcloud-resource-registry.yaml to deploy the overcloud on their own --
without having to do any pre-processing (via merge.py).
To test these you have to add the resource registry environment (in
addition to the existing `-e` option) and use the new overcloud template
in the Heat call in devtest_overcloud.sh (line 374):
heat $HEAT_OP -e $TRIPLEO_ROOT/overcloud-env.json \
-e "$TRIPLEO_ROOT/tripleo-heat-templates/overcloud-resource-registry.yaml" \
-t 360 \
-f $TRIPLEO_ROOT/tripleo-heat-templates/overcloud-without-mergepy.yaml \
-P "ExtraConfig=${OVERCLOUD_EXTRA_CONFIG}" \
$STACKNAME
The existing overcloud Heat environment
($TRIPLE_ROOT/overcloud-env.json) should keep on working. Scaling is
now being controlled by the `ControllerCount` and `ComputeCount`
template parameters, though.
NOTE: the changes here depend on a fairly recent Heat build (commit
e5f285f6cb from ~7th September, 2014). In other words, this requires
Juno Heat.
Also, passing more than one environment file to Heat requires
python-heatclient version 0.2.11.
Change-Id: I687a00c7dc164ba044f9f2dfca96a02401427855
|
Dan Prince
Steven Hardy
Jenkins
Giulio Fidente
Steve Baker
Dan Sneddon
Emilien Macchi
Peter Belanyi
Tomas Sedovic