| Age | Commit message (Collapse) | Author |
|---|
|
Add gluster 3.9 templates
|
|
|
|
Add in-tree CI scripts
|
|
Currently, the scripts that contain testing inventory
and actually execute the running of openshift-ansible
plays are kept in the CI jobs themselves.
This commit adds scripts into our tree so we can have
better control over variables, differences between branches,
and more transparency to end users for how we run our
integration tests.
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
mgugino-upstream-stage/cast-bool-openshift_docker_use_system_container
Automatic merge from submit-queue.
Cast openshift_docker_use_system_container to bool
openshift_docker_use_system_container might be passed
in via ini inventory as 'openshift_docker_use_system_container=false'
This condition will be interpreted as a string type, instead of
boolean.
Casting openshift_docker_use_system_container as bool
will achieve the users desired intent.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1528943
|
|
Automatic merge from submit-queue.
Add docker auth credentials to system container install
This commit adds docker auth credentials mount to
system container systemd unit file.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1514324
|
|
Automatic merge from submit-queue.
Remove oauth_template bits from openshift_facts
This commit moves some deprecated variable logic
outside of openshift_facts and into role defaults.
|
|
Automatic merge from submit-queue.
Correct kublet_args cloud-provider directories
This commit updates kublet_args to correct
cloudprovider configuration file locaitons.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1527203
|
|
Automatic merge from submit-queue.
Move wait_for_pods to it's own play openshift_hosted
Currently, both registry and router pods need to
be polled for successful deployment.
Somtimes this can take up to a minute.
This commit attempts to deploy both pods
before polling either. This should reduce
the average wait time for polling pods by 50%
as time spent polling the first will also allow
the second pod to continue it's own deployment.
|
|
Automatic merge from submit-queue.
Update logging to use existing cluster deployment for defaults
This will allow us to use logging facts to set defaults of specific configurations such as ES index replicas and shard count.
The update to logging facts yields us output like:
```json
"elasticsearch": {
"clusterrolebindings": {},
"configmaps": {
"logging-elasticsearch": {
"elasticsearch.yml": {
"cloud": {
"kubernetes": {
"namespace": "${NAMESPACE}",
"pod_label": "${POD_LABEL}",
"pod_port": 9300
}
},
"cluster": {
"name": "${CLUSTER_NAME}"
},
"discovery": {
"type": "kubernetes",
"zen.minimum_master_nodes": "${NODE_QUORUM}",
"zen.ping.multicast.enabled": false
},
"gateway": {
"expected_nodes": "${RECOVER_EXPECTED_NODES}",
"recover_after_nodes": "${NODE_QUORUM}",
"recover_after_time": "${RECOVER_AFTER_TIME}"
},
"index": {
"number_of_replicas": 0,
"number_of_shards": 1,
"translog": {
"flush_threshold_period": "5m",
"flush_threshold_size": "256mb"
},
"unassigned.node_left.delayed_timeout": "2m"
},
"io.fabric8.elasticsearch.authentication.users": [
"system.logging.kibana",
"system.logging.fluentd",
"system.logging.curator",
"system.admin"
],
```
TODO:
- [x] Update logging facts to pull out settings from config maps
- [x] Move `openshift_sanitize_inventory/library/conditional_set_fact.py` up to repo level
- [x] Generate diffs against currently deployed configs and correctly patch in custom changes from customers
- [x] Use `conditional_set_fact` to easily set defaults for logging based on logging facts, or falling back to role defaults when not specified in the inventory
- [x] Update all components to follow patching configmaps
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
Automatic merge from submit-queue.
Bug 1527178 - installation of logging stack failed: Invalid version s…
…pecified for Elasticsearch
openshift_logging_{curator,elasicsearch,fluentd,kibana,mux}/vars/main.yml:
- adding "3_8" to __allowed_.*_versions
- replacing the value of __latest_.*_version "3_6" with "3_8".
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1527178
|
|
openshift_docker_use_system_container might be passed
in via ini inventory as 'openshift_docker_use_system_container=false'
This condition will be interpreted as a string type, instead of
boolean.
Casting openshift_docker_use_system_container as bool
will achieve the users desired intent.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1528943
|
|
This commit updates kublet_args to correct
cloudprovider configuration file locaitons.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1527203
|
|
Remove bootstrap.yml from main.yml in openshift_node role
|
|
files, use diffs to keep custom changes, white list certain settings when creating diffs
|
|
This commit adds docker auth credentials mount to
system container systemd unit file.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1514324
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
specified for Elasticsearch
openshift_logging_{curator,elasicsearch,fluentd,kibana,mux}/vars/main.yml:
- adding "3_8" to __allowed_.*_versions
- replacing the value of __latest_.*_version "3_6" with "3_8".
|
|
Automatic merge from submit-queue.
Adding no_log to registry_auth.
We were seeing credentials show up in our jenkins logs. We'd prefer if they were not logged. I'm not sure the repercussions but these are sensitive and probably should be `no_log`.
|
|
Automatic merge from submit-queue.
Laying config files down in a daemonset pod.
Work to have daemonset config laydown files for nodes used with golden images.
Based on the original work from @smarterclayton https://github.com/openshift/origin/pull/17351
|
|
Currently, both registry and router pods need to
be polled for successful deployment.
Somtimes this can take up to a minute.
This commit attempts to deploy both pods
before polling either. This should reduce
the average wait time for polling pods by 50%
as time spent polling the first will also allow
the second pod to continue it's own deployment.
|
|
Automatic merge from submit-queue.
Must directly specify google-cloud-sdk version
The upstream forces versions on packages, and rpm -V fails when it does
not match. Also add which, used by gcloud.
Fixes break https://ci.openshift.redhat.com/jenkins/job/build-and-release-latest-openshift-ansible-images/210/console
|
|
Automatic merge from submit-queue.
aws: Fix misnamed variable in provisioning_vars.yml.example
This typo (?) in `provisioning_vars.yml.example` tripped me up while trying to run `provision_install.yml` using a configuration based on the example file.
|
|
This commit utilizes include_role for bootstrapping the
node instead of conditional include of tasks now that
the node role has no meta includes that have tasks.
|
|
Move sanity_checks into custom action plugin
|
|
Fix container_runtime openshift_containerized_host_groups
|
|
|
|
openshift_containerized_host_groups needs to be referenced via
hostvars.
This commit also updates tox ansible syntax checks to account
for unavailability of hostsvars during syntax checks.
Fixes: https://github.com/openshift/openshift-ansible/issues/6540
|
|
Move openshift_deployment_type check into sanity_check
action plugin. Remove compatibility for deployment_type.
deployment_type has been deprecated for some time now.
|
|
The upstream forces versions on packages, and rpm -V fails when it does
not match. Also add which, used by gcloud.
|
|
|
|
This commit moves some deprecated variable logic
outside of openshift_facts and into role defaults.
|
|
Automatic merge from submit-queue.
Add gcloud to the installer image
The installer image will become our default cloud operations image for
tasks that overlap with ansible, and will also be the image for
provioning. We will eventually add others like awscli for support roles.
Prereq for moving origin-gce into this repo
@sdodson as discussed earlier (bypasses the need to get this in our tree for now)
|
|
Move more checks outside of init/main.yml for
speeding up upgrades and other operational plays that
need to run.
|
|
This commit moves sanity_checks tasks into a custom
action plugin that is only run against a single host.
This will result in a large reduction of tasks during initialization
|
|
Automatic merge from submit-queue.
Remove openshift.common.{is_atomic|is_containerized}
We set these variables using facts in init, no need
to duplicate the logic all around the codebase.
|
|
We set these variables using facts in init, no need
to duplicate the logic all around the codebase.
|
|
Adding support for docker-storage-setup on overlay
|
|
Automatic merge from submit-queue.
Move node group tags to openshift_aws_{master,node}_group.
Move node groups tags to `openshift_aws_{master,node}_group` variables as discussed below.
|
|
|
|
Add readiness probe to kuryr controller pod
|
|
The installer image will become our default cloud operations image for
tasks that overlap with ansible, and will also be the image for
provioning. We will eventually add others like awscli for support roles.
|
|
Add CentOS-OpenShift-Origin37 repo template.
|
|
Plugin consolidate
|
|
Automatic merge from submit-queue.
Remove some small items from openshift_facts
Remove some small items from openshift_facts that
don't need to be there.
|
|
Automatic merge from submit-queue.
Fix hosted_reg_router selectors
This commit removes duplicate (and incorrect) assignment
of openshift_hosted_registry_selector and
openshift_hosted_router_selector
Fixes https://github.com/openshift/origin/issues/17682
|
