aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2017-01-11Merge "Update net-config-noop to use apply-config"Jenkins1-2/+2
2017-01-11Merge "Removes deprecated OpenDaylight L2 only deployments"Jenkins2-6/+3
2017-01-11Merge "Add support for the deployment of Ceph MDS"Jenkins4-0/+53
2017-01-11Fix error path for tools/process-templates.pySteven Hardy3-0/+4
The error path for this fails because we don't import six or install it in the templates tox venv Change-Id: Ie9f46332f2b03d48a1b0a4a432e9721757833569
2017-01-11Allow freeipa environment file to be in /tmpJuan Antonio Osorio Robles1-0/+2
Relying on home can be tricky if run on CI, since we would need to hard-code the user. So for this case, we can use /tmp to source that environment file from. bp tls-via-certmonger Change-Id: Ie6234a15af7aa493257c13bd40cbf51ab36085be
2017-01-10Merge "Add example showing how to set root password via cloud-init"Jenkins1-0/+38
2017-01-10Removes deprecated OpenDaylight L2 only deploymentsTim Rozet2-6/+3
Deploying ODL without L3 DVR is no longer supported. This patch moves the opendaylight-l3 env settings to be the new default for opendaylight env file, while also removing any option to disable L3. Closes-Bug: 1654586 Change-Id: Ia2488cb0b752fb4b33f03caa7a1d2469d20395f0 Signed-off-by: Tim Rozet <trozet@redhat.com>
2017-01-10Merge "Updated from global requirements"Jenkins3-4/+17
2017-01-10Merge "README: Add mistral"Jenkins1-0/+2
2017-01-10Add deployed-server backwards compatible templateJames Slagle4-2/+35
In Newton, the ctlplane port on deployed-server was called <hostname>-ctlplane-port. When this code was refactored in I29fbc720c3d582cbb94385e65e4b64b101f7eac9, the -port suffix was dropped in favor of <hostname>-<network> convention, and the port resource was created directly in deployed-server.yaml instead of in a nested stack. Both of those changes were backwards incompatible -- making it impossible to upgrade to the new version of deployed-server.yaml without the ctlplane port getting deleted/recreated, which causes a change in IP address. The IP address change causes services to be misconfigured on upgrade attempts. Change-Id: I45991b60a151abf3c5e4d05a3aa7246b2d25ac5a
2017-01-10Merge "Setting networking-odl v2 as a mechanism driver"Jenkins2-3/+3
2017-01-10Add support for the deployment of Ceph MDSGiulio Fidente4-0/+53
This change adds a CephMds service, disabled by default, on the Controller role and an environment file to enable it. Change-Id: If7cb46319038a80ed52f753a623989885e1b7da4 Depends-On: Iaecc3ff7acb851776c5057c42a5a513a70425d2c Partial-Bug: #1644784
2017-01-10Add THT for networking-fujitsuKoki Sanagi5-0/+101
Introduce THT for networking-fujitsu. networking-fujitsu is a neutron ML2 plugin which enables FUJITSU C-Fabric switch in OpenStack environment. This templates deploy overcloud with C-Fabric switch. Change-Id: Iee75a1a30552d8dc9f55f52d10b0dc2b623992ef Implements: blueprint integration-networking-fujitsu Depends-On: I37a502b43eb7d91bfe20625248ed117eae3ca535 Depends-On: I5eb2c2a9c50b5991d62f4b6d74b83351c86b02de
2017-01-10Add retry to RHEL registrationCharles Llewellyn1-13/+30
Occasionally we can see transient network outages when attempting to register with the Redhat Portal or Satellite server. This causes deployment or scaleout operations to fail. These outages are minimal and retrying often resolves the issue. This becomes more prevelant during testing as we deploy infrastructure far more frequently. Change-Id: If23785fbe2eea4643918b2e68915bbc13c1b1112
2017-01-10Updated from global requirementsOpenStack Proposal Bot3-4/+17
Change-Id: I9ae60dfb2ef5e8fb5aae482ddb08206fc1980ca5
2017-01-10Merge "Make update-from-keystone-admin-internal-api.yaml work on newton+"Jenkins1-28/+1
2017-01-09Merge "Configure Kernel Args and Tuned and then reboot for Compute"Jenkins3-0/+174
2017-01-09README: Add mistralEmilien Macchi1-0/+2
Mistral is tested by scenario003. Let's document it. Change-Id: If715a2071b7e421b63033fd00f49d3a3f1948af1
2017-01-09Configure specific module path for ansible.Sofer Athlan-Guyot1-0/+1
Change-Id: Icd23553e94fd573a8032c52e18857e722eb2ea49
2017-01-09Merge "Add FreeIPA server installation script"Jenkins1-0/+94
2017-01-09Merge "Template and role support for the undercloud"Jenkins9-11/+366
2017-01-09Merge "scenario001: add Panko testing"Jenkins2-0/+4
2017-01-09Setting networking-odl v2 as a mechanism driverItzik Brown2-3/+3
The current environment files for OpenDaylight use the opendaylight mechanism driver - changing it to opendaylight_v2. Also using the l3_odl_v2 as a service plugin Depends-On: I2a1c5097614e47cc09e43bbc77305a0548d54baa Change-Id: Iba6f52663ce4a5e4c070f58f079465fdd24ecc7c
2017-01-09Merge "Add missing VIP definitions into -no-tunneling env file"Jenkins2-6/+27
2017-01-09Make update-from-keystone-admin-internal-api.yaml work on newton+Cyril Lopez1-28/+1
There are change of ServiceNetMapDefaults in service_net_map.j2.yaml but were not reproduce in update-from-keystone-admin-internal-api.yaml environment. Tested in newton. Closes-Bug: #1646862 Change-Id: I307dcaabbc6d583896090bf3f046b442007fbc42 Signed-off-by: Cyril Lopez <cylopez@redhat.com> Co-Authored-By: Gregory Charot <gcharot@redhat.com>
2017-01-08Continue checking for request url if initially nullJames Slagle1-1/+8
When using get-occ-config.sh during overcloud upgrades, the script could potentially be started before the stack has been upgraded. In that scenario, the script will return null for the request metadata url for the deployed-server resource since the stack has not yet been updated and it's still using the previous Heat signaling. This patch updates the script to just continue checking the resource metadata if the request url is null. Eventually, once the resource has been updated, the script will continue and properly populate the os-collect-config configuration. Change-Id: I9db54d8ad278715f42b768edf8f0fd21998b2098
2017-01-08Update net-config-noop to use apply-configDan Prince1-2/+2
Some deployments of the undercloud would like to use net-config-noop (to disable all network configuration). This patch changes this noop software deployment to use 'apply-config' instead of 'os-apply-config' which allows it to work with the undercloud installer which only supports the newer (apply-config) style due to heat signalling requirements. Change-Id: If7b579b759477afd81ad64a4847169e59ddcbd77
2017-01-08Add UpgradeInitCommand to deployed-serverJames Slagle1-0/+26
The commands specified by UpgradeInitCommand need to be run before InstanceIdDeployment in deployed-server.yaml, otherwise the upgrades hang with the resource in progress. This is because the new python-heat-agent-apply-config has not yet been installed on the deployed server. Adding the UpgradeInitCommand (and corresponding SoftwareConfig/SoftwareDeployment to apply it) will cause the new repos and python-heat-agent-* rpm's to be installed before InstanceIdDeployment. An open question is whether or not Heat should even be triggering the InstanceIdDepoyment to IN_PROGRESS on upgrade when only the group is changing from os-apply-config to apply-config. If that turns out to be a Heat bug, then this patch wouldn't be necessary. Change-Id: I9d87f995744415b110a7d0bca8d2309d7167148c
2017-01-06Template and role support for the undercloudDan Prince9-11/+366
Add a new roles data YAML file and environment to help create the undercloud via t-h-t. Partially-implements: blueprint heat-undercloud Change-Id: I36df7fa86c2ff40026d59f02248af529a4a81861
2017-01-06Heat engine settings required for UndercloudDan Prince1-0/+12
This patch adds several new heat engine setting that are required in order to configure a fully working Heat engine in the undercloud. New parameters have been added for the max resources and convergence settings. I've hard coded the max stack depth at 6 which matches our undercloud and should be fine as a default setting I think (would rarely need overridden). Change-Id: I335476f46b4bea4b9f3ac70a6312a7e906f43ba6
2017-01-06Merge "Set rabbitmq package_provider to yum"Jenkins1-0/+1
2017-01-06Merge "Add custom roles data for deployed-server"Jenkins2-0/+176
2017-01-06scenario001: add Panko testingEmilien Macchi2-0/+4
Panko has been introduced in TripleO but is not tested yet. Adding it on scenario001 with other telemetry services. Depends-On: If2f6559a7d76b26fa9b0a3ecfa2e2101aae93e3c Change-Id: Ic76f1cb36ee05e22cfda1fff8ad3f0ec18acf617
2017-01-06README: updates for CI matrixEmilien Macchi1-2/+2
Update the CI matrix to fit with what is actually tested. Change-Id: Ib9c96fe7a3ff25043395cb5a89657b3b6d22f12a
2017-01-06Remove unused pacemaker profilesMichele Baldessari39-1651/+0
With change I80c8559bb2d915385bcc20ae71fe144ddd6591c1 we removed all the unused puppet-tripleo pacemaker profiles. With this change we remove the corresponding puppet profiles from tripleo-heat-templates. We can also remove any trace of the fake ::Core service as it was introduced via Iacd94294b8a66bc082bb2b3e8d3364ec1bf053b8 for the fake openstack-core pacemaker resource during the Mitaka cycle and became unused in Newton. Change-Id: I48cd2b6a4593d673d5883b45feae088392e7e713
2017-01-06Configure Kernel Args and Tuned and then reboot for ComputeSaravanan KR3-0/+174
* On top of the https://review.openstack.org/#/c/411204 * Added Kernel args and Tune-d configuration * Added provision to provide different kernel args per role (applicable for different types of compute roles only) Implements: blueprint tuned-nfv-dpdk Change-Id: I5c538428c376c9d2ebd1c364f0ee8503fd7d620e
2017-01-05Merge "Merge mysql service_config_settings for nova"Jenkins1-15/+17
2017-01-05Merge mysql service_config_settings for novaJames Slagle1-15/+17
The hieradata from the nova-base.yaml service template needs to be available to the role running mysql, which isn't necessarily the same role as nova. nova-base.yaml isn't an actual service template though that is included in any ServiceChain resources, it's outputs need to be merged with an actual nova service template, such as nova-api.yaml. As nova-api.yaml already provides some hieradata for the mysql service in service_config_settings, this patch uses map_merge to combine the 2 values. Change-Id: I4dc684b3611b13f177f9499e69468d3f6ef6fa76 Closes-Bug: #1654058
2017-01-05Merge "nova-api: legacy cleanup with old wsgi params"Jenkins1-11/+0
2017-01-05Merge "nova-api: also include hiera for new apache_api class"Jenkins1-3/+14
2017-01-05Merge "Configure Heat engine metadata URL's"Jenkins1-0/+17
2017-01-04puppet/role.role.j2.yaml has invalid get_resource referenceDan Prince1-1/+1
Found this today when rebasing the undercloud installer. The puppet/role.role.j2.yaml Yaml has an invalid get_resource reference that causes a cryptic heat stack failures. Change-Id: Icfb7d73a1c4d02213b23a427605f2b0d5eaa984f
2017-01-04nova-api: legacy cleanup with old wsgi paramsEmilien Macchi1-11/+0
Cleanup old legacy params for wsgi config. Change-Id: Ic775de171c95d43d9273e1a29db2ab685fdf7706 Depends-On: I59b3b36be33268fa6e261a7db3c4aa8e8e712ffb
2017-01-04nova-api: also include hiera for new apache_api classEmilien Macchi1-3/+14
puppet-nova renamed nova::wsgi::apache to nova::wsgi::apache_api to welcome nova::wsgi::apache_placement (for nova placement API). This patch adds the required parameters before we make the switch in puppet-tripleo. Legacy parameters will be removed when the switch will be done in puppet-tripleo. Change-Id: I5fc99062d349597393e2248c66f2d863029c7730
2017-01-04Merge "Add pre-network hook and example showing config-then-reboot"Jenkins8-0/+85
2017-01-04Set rabbitmq package_provider to yumJames Slagle1-0/+1
When deploying with EnablePackageInstall:True, the rabbitmq puppet module defaults to the rpm package provider, which then tries to "rpm -i undef" since we are setting rabbitmq::package_source to undef. Instead of using the rpm provider at all, we should just use the yum provider to install whatever rabbitmq rpm's are found in enabled repos. Change-Id: I29365e675bfde676fde7a54dfc6c660c3970f50a Partially-implements: blueprint split-stack-software-configuration
2017-01-04Add custom roles data for deployed-serverJames Slagle2-0/+176
Adds a custom roles data file for use when using the deployed-server templates. The file takes care of setting disable_constraints: True, so that deployers don't have to do things like create fake images in glance. Also adds a comment to roles_data.yaml documenting disable_constraints. Partially-implements: blueprint split-stack-software-configuration Change-Id: I7c26c0c2851e0d6bcea42d7af7f4295a1944ec9f
2017-01-04Merge "Adds missing firewall rules for OpenDaylight API service"Jenkins1-0/+6
2017-01-04Add example showing how to set root password via cloud-initSteven Hardy1-0/+38
There have been some requests to enable this, and although it's not something we should enable by default, this shows how you can use the generic NodeUserData interface to have cloud-init do it. To use this you create an environment file like: resource_registry: OS::TripleO::NodeUserData: path/to/userdata_root_password.yaml parameter_defaults: NodeRootPassword: insecure_changeme Obviously this isn't that secure, and thus isn't intended for production environments, but it may be useful for debugging and as a further example showing how to configure things via cloud-init. Change-Id: If87a1e1dbfaf31b84cc0667c9a60bbd3c757d8cd
2017-01-04Merge "DB connection: prevent src address from binding to a VIP"Jenkins19-0/+83