aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2016-05-19composable heat servicesBrad P. Crochet15-220/+378
Adds new puppet and puppet pacemaker specific services for Heat API, Heat API CFN, Heat API Cloudwatch, and Heat Engine. The Pacemaker templates extend the default heat services and swap in the pacemaker specific puppet-tripleo profile instead. Change-Id: I387b6bfd763d2d86cad68a3119b0edd0caa237b0 Partially-implements: blueprint composable-services-within-roles Depends-On: I194cbb6aa307c2331597147545cf10299cab132f Depends-On: I14dc923ac8ee8d5d538e7f4cf8138ccee8805b53
2016-05-19Deploy Loadbalancer as a composable roleEmilien Macchi7-78/+63
Deploy loadbalancer service using puppet-tripleo, and drop puppet code. Implements: blueprint refactor-puppet-manifests Depends-On: I9b106dcc1a4d446ab5dea8430ed295e6ec209cbd Change-Id: I9ca50a4bc822ec17d89988894af9bdf07e4bd1a9
2016-05-19Merge "Add net-config-static.yaml"Jenkins1-0/+82
2016-05-19Merge "Remove ControllerClusterConfig from overcloud.yaml"Jenkins2-30/+0
2016-05-19Merge "add heat-api-cfn to endpoint map"Jenkins3-0/+138
2016-05-19Merge "Add missing ManagementIpSubnet"Jenkins1-0/+4
2016-05-19Merge "Change ping wait flag"Jenkins1-1/+1
2016-05-19Tighten the access rules for galeraMichele Baldessari3-11/+68
Set a password for the 'root' db user and add an additional 'clustercheck' user to be used only by the resource agent. The password for this 'clustercheck' user is randomly generated via a heat parameter. Before this change the workflow to set up the database in the manifest is the following: - Step 1 -> Install all the basic galera packages and basic configuration - Step 2.a -> Create /etc/sysconfig/clustercheck with root and empty password - Step 2.b -> Start up galera-monitor xinetd service - Step 2.c -> Start pacemaker ocf resource (no root user has been created so there will be an empty password per default) - Step 2.d -> Wait for /bin/clustercheck to return success and then proceed with the other steps After this change the workflow is slightly more complex because there is a bit of a chicken and egg problem: - Step 1 -> Install all the basic galera packages and basic configuration - Step 2.a -> Create /etc/sysconfig/clustercheck with root and empty password unless the file does exists already and has a clustercheck user configured - Step 2.b -> Start up galera-monitor xinetd service - Step 2.c -> Start pacemaker ocf resource (no root user has been created yet, so there will be an empty password per default) - Step 2.d -> Wait for /bin/clustercheck to return success and then proceed with the other steps - Step 2.e -> Create clustercheck db user - Step 3/4 -> Create /etc/sysconfig/clustercheck with clustercheck user credentials - Step 5.a -> Update the sql root password on the each node (at this stage - Step 5.b -> Create /root/.my.cnf with proper credentials on all nodes Note that we cannot really create the root/clustercheck users right at step 1 because the db is not running yet (an approach that spawned mysqld on each node, created the users and shut it down, was tried but was much more complex and cannot work on updating existing setups) Given the new way of solving the root password issue, we also need to make sure that Step1 and Step2 are running on updates. Closes-bug: #1581677 Depends-On: I83eed8885503043e881db34411616f9726e00352 Change-Id: If3d6e7253af6195b96129be7ea3348d697e4bae1
2016-05-18Deploy RabbitMQ as a composable roleEmilien Macchi8-94/+78
Change the way to implement RabbitMQ, as a composable role. Implements: blueprint refactor-puppet-manifests Change-Id: I5fed5c437ad492af75791a9163f99ae292f58895
2016-05-18Merge "Add step to ObjectStorage RingBuilder deployment"Jenkins1-0/+5
2016-05-18Dump IPs configuration as hieradataGiulio Fidente8-76/+105
This might be useful if we switch to %{hiera()} calls to lookup the bind address from within a service. Also gets rid of NetIpSubnetMap and provides same output from NetIpMap instead. Change-Id: I328a417d1f1fff9c31e9ad7b2b5083ac19bc7329
2016-05-18Pass enabled and manage_service to Neutron/L3 when PacemakerGiulio Fidente1-0/+2
We can drive enabled and manage_service from the Pacemaker specific templates so change I8ef7bb94e048b998712b3534ceb51a7d10d016e9 removed that logic from the TripleO specific manifests. This change adds them into the Pacemaker template for the Neutron/L3 agent. Change-Id: I1146a8bae4938577e9927d1ce74009e066cb58dd
2016-05-18Add step to ObjectStorage RingBuilder deploymentSteven Hardy1-0/+5
https://review.openstack.org/#/c/236243 added a new conditional for the controller steps, but we don't pass any step for the ObjectStorage nodes, so the deployment fails. This passes a step that enables the ringbuilder again, although it does end up inconsistent with the deployment Step name. Change-Id: I506961f4a22dba9960d819d7376a39e7ccbcdece Closes-Bug: #1583225
2016-05-18composable neutron metadata serviceDan Prince11-27/+86
Adds new puppet and puppet pacemaker specific services for the Neutron Metadata agent. Partially-implements: blueprint composable-services-within-roles Change-Id: I25f026507e78f18594599b3621613a54f246545d
2016-05-18composable neutron l3 serviceDan Prince12-43/+79
Adds new puppet and puppet pacemaker specific services for the Neutron L3 agent. Partially-implements: blueprint composable-services-within-roles Change-Id: I0316043efe357a41ef3b4088a55d98dbb6d25963
2016-05-18Merge "Ensure manage_service and enabled are false for Pacemaker"Jenkins4-3/+10
2016-05-17Move dhcp agent maps to OS::Heat::None to resource_registryJames Slagle2-3/+4
Change Id Ia61295943e67efe354a51a26fe4540f288ff6ede added support for composable Neutron dhcp agent services. However in the sample environments for OpenContrail and Plumgrid where dhcp agent is disabled, the mapping for OS::Heat::None was under parameter_defaults instead of resource_registry. Change-Id: I0aedbbc3720783d4208d524cd28c7eed4fc5d1d7
2016-05-17Remove ControllerClusterConfig from overcloud.yamlSteven Hardy2-30/+0
AFAICS this isn't actually used anywhere, I assume it's left over from the older element based implementation. Change-Id: Ie95628bd7af1bcd50a6e331531b2987e434c7136
2016-05-16Merge "Remove Nova EC2 deployment"Jenkins5-147/+0
2016-05-16Remove tenant_id from nova v2.1 endpointBen Nemec2-6/+6
Per the nova devs on [1], this is not necessary. Change-Id: I11974432c995b22b3c98ef9ae2adc3508d9cc536 1: https://review.openstack.org/#/c/316241/1/manifests/keystone/auth.pp
2016-05-16Remove Nova EC2 deploymentEmilien Macchi5-147/+0
Nova EC2 does not exist anymore since Mitaka, parameters are already deprecated in Mitaka and send warnings to the Puppet catalog. The service has been replaced by ec2api project, where Puppet OpenStack team is currently writting a module. In the meantime we add support in TripleO, this patch removes all occurences of Nova EC2 configuration, which are useless and send warnings for nothing. Change-Id: Ief2d0e5c77b5ac58560606fee930fbd66c40ffc3
2016-05-16Merge "Add database directory mount for openvswitchdb"Jenkins1-0/+1
2016-05-13Ensure manage_service and enabled are false for PacemakerGiulio Fidente4-3/+10
We can control the two manage_service and enabled boolean from the Pacemaker specific template now. Change-Id: I91a4267f0fc230f63df3333747d28463c7ae55fe
2016-05-12add heat-api-cfn to endpoint mapSteven Hardy3-0/+138
Change-Id: I8f98ce92fc387d2263fda738c1c8a209e3cbbb85
2016-05-11Merge "composable neutron dhcp service"Jenkins12-194/+139
2016-05-10composable neutron dhcp serviceDan Prince12-194/+139
Adds new puppet and puppet pacemaker specific services for the Neutron DHCP agent. Depends-On: Ibbfd79421f871e41f870745a593cca65e8c0e58a Partially-implements: blueprint composable-services-within-roles Change-Id: Ia61295943e67efe354a51a26fe4540f288ff6ede
2016-05-10deployment: drop step6Emilien Macchi2-13/+6
Step6 was just about confuring fencing after creating all Pacemaker resources. It was created by this patch: https://review.openstack.org/#q,1787fbc7ca58f9965cd5d64b685c1f9beed4cb9b,n,z A bit of Puppet orchestration can help us to not require an extra step. This patch: * configure & enable fencing at step5 * make sure we don't configure fencing because creating Pacemaker resources and constraints. * remove step6 from deployment workflow. * depends on a patch in puppet-tripleo that moves keystone resources (endpoints, roles) to step 5. Change-Id: Iae33149e4a03cd64c5831e689be8189ad0cf034b Depends-On: Icea7537cea330da59fe108c9b874c04f2b94d062 Depends-On: I079e65f535af069312b602e8ff58be80ab2f2226
2016-05-10deployment: remove Step7Emilien Macchi1-12/+1
Step7 was created when we incremented the step of ringbuilder, by https://review.openstack.org/#q,9988bd25aa4bac1375ef4783d636c7adecedee92,n,z But step7 is not used anywhere and consumes some times for nothing. This patch removes the step, so deployments and upgrades will be faster. Change-Id: I77af9126abc61ace227cf1a69c2d3b5ceb735276
2016-05-10Set nova neutron auth back to 'v3password'.Derek Higgins1-0/+2
Puppet-nova recently changed the default neutron auth setting in I3416ae594e972e40ff0336779258a887987e46b1 to 'password'. This single setting seems to break the tripleo upgrades job. Setting it here manually for now and following up in puppet-nova. Closes-bug: #1580076 Change-Id: I3f38a3e1ef3378a272a51ecbc1e8a801c8d3608a
2016-05-09Change ping wait flagFeng Pan1-1/+1
This is an optimization of the ping command. It changes ping test from waiting for a single sent packet for 300 seconds to waiting to receive a packet from any number of sent packets. The current implemenation waits for full 300 seconds before retrying if we do not get a reply for the first packet sent. By using -w flag, we keep sending icmp packets until a reply is sent, making the ping test more responsive to connectivity change. Change-Id: I01ab374ae44718c8d56e2d7f35812dfb5bb2ce5a Signed-off-by: Feng Pan <fpan@redhat.com>
2016-05-09Merge "Pass parameters to manage endpoints via puppet"Jenkins3-0/+103
2016-05-06Use docker-cmd hook.Ian Main2-4/+4
This patch switches to use docker-cmd without changing the heat templates. Change-Id: I4a6a42819e83e3b70bf1e37c09d155c5cf8a7ee4
2016-05-06Merge "Document IPv6 syntax for GlanceFilePcmkDevice"Jenkins1-0/+2
2016-05-06Merge "Fixes ovs bonding option"Jenkins1-1/+1
2016-05-06Merge "Additional parameters for Nuage Neutron plugin integration"Jenkins1-0/+9
2016-05-05Merge "Remove calls to ::mysql from the manifests"Jenkins2-8/+0
2016-05-05Combine parameter_default fieldsRyan Hallisey1-2/+0
The paramaters field was changed to parameter_defaults everywhere. Combine both parameter_default fields. Change-Id: Ia1874463cdd6bf81be5739363e639bd11312abec
2016-05-04Pass parameters to manage endpoints via puppetJuan Antonio Osorio Robles3-0/+103
This commit passes the necessary hieradata in order to create the endpoints, users and roles of the services in keystone via puppet. Change-Id: I2470dfa4661be7ba8218f6035fffa05f547214f0
2016-05-04Leave start/stop/restart for Keystone and Glance in charge to the roleGiulio Fidente1-2/+0
Change-Id: I511052dc765788336ffd32dee2118d787fce725d
2016-05-04Remove calls to ::mysql from the manifestsGiulio Fidente2-8/+0
The database will be created by the roles so we don't need to call ::mysql from the manifest. Change-Id: I2b137cbd6597222a72cf46830f34a93f002c70ef Depends-On: Id065a9180f1f1a41ab225ec5f755498ec7d9a827
2016-05-04Merge "Disable VIPs before stopping cluster during version upgrade"Jenkins2-0/+11
2016-05-04Merge "Fix distinguishing between stack-create and stack-update"Jenkins3-5/+8
2016-05-04Merge "Use str_split to compute netmask in _v6 port templates"Jenkins11-44/+22
2016-05-03Optimize Nova disk_cachemodes and hw_disk_discard options for RBDGiulio Fidente1-2/+11
When using the Nova RBD driver for the ephemeral storage it is suggested by the Ceph RBD OpenStack guide [1] to optimize certain settings; this change will set disk_cachemodes and hw_disk_discard accordingly to the guide. 1. http://docs.ceph.com/docs/master/rbd/rbd-openstack/ Change-Id: I8d2ee89ca4ff5458d1888cc037e2e91d19025ad4
2016-05-03Merge "Wire missing RabbitClientPort into Glance API role"Jenkins1-0/+5
2016-05-03Merge "Updates ControlPlaneSubnetCidr to be a string"Jenkins1-1/+1
2016-05-02Disable VIPs before stopping cluster during version upgradeIan Pilcher2-0/+11
If "pcs cluster stop --all" is executed on a controller that happens to have a VIP on the internal network, pcs may use the VIP as the source address for communication with another cluster node. When pacemaker is stopped this VIP goes away, and pcs never receives a response from the other node. This causes pcs to hang indefinitely; eventually the upgrade times out and fails. Disabling the VIPs before stopping the cluster avoids this situation. Change-Id: I6bc59120211af28456018640033ce3763c373bbb Closes-Bug: 1577570
2016-04-29Use str_split to compute netmask in _v6 port templatesGiulio Fidente11-44/+22
Change-Id: I282dbc025500b1628d4f08a49b54a2adefd38b5f
2016-04-29Merge "Fix the Sahara public endpoint and make it point to the Public VIP"Jenkins2-4/+4
2016-04-29Merge "Fix ControllerExtraConfig parameter name"Jenkins2-5/+24