| Age | Commit message (Collapse) | Author |
|
|
|
Bootstrap enhancements.
|
|
Currently, we check that upgrade target is less than 3.8,
but this will break for minor upgrades.
This commit set's a fact early in the upgrade process to
deterime what the currently installed version on the
first master is.
This fact is used to determine if our currently installed
version is less than 3.7.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1508301
|
|
This also causes a failure if nfs block is not defined, because the condition is missing.
Anyway this shouldn't have been committed.
|
|
|
|
Automatic merge from submit-queue.
Temporarily set master servingInfo.clientCA as client-ca-bundle.crt during rolling CA redeployment.
This PR temporarily configures master `servingInfo.clientCA` as `client-ca-bundle.crt` during CA redeploy so that clients referencing the old OpenShift CA can continue to function until all cluster certificates have been replaced and `servingInfo.clientCA` is reset to `ca.crt`.
`client-ca-bundle.crt` is a new bundle created by the `openshift_ca` role which contains old and new OpenShift CA certificates.
https://bugzilla.redhat.com/show_bug.cgi?id=1509124
Ref: https://bugzilla.redhat.com/show_bug.cgi?id=1493276
|
|
mgugino-upstream-stage/retry-restart-master-controllers
Automatic merge from submit-queue.
Retry restarting master controllers
Currently, master controller services may fail to restart
if master api services are not fully initialized.
This commit enables retry of master controllers.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1509837
|
|
rolling CA redeployment.
|
|
Currently, master controller services may fail to restart
if master api services are not fully initialized.
This commit enables retry of master controllers.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1509837
|
|
Currently, this task is executed based on openshift_version.
openshift_version is based on the upgrade target, thus not
the currently install versions.
This commit ensures that the task executes as intended.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1508301
|
|
|
|
Automatic merge from submit-queue.
management: users must acknowledge use of beta software
Adds a new role variable, openshift_management_install_beta. This
variable defaults to false. The value of this variable is checked
during the validation phase.
* If true, the install will not continue.
* If false, The user is presented with an informative message letting
them know this is beta software and there is low/no support at this
time. The installation will abort and instruct the user how to
continue.
|
|
Automatic merge from submit-queue.
Bootstrap enhancements.
This includes the elb work.
|
|
Adds a new role variable, openshift_management_install_beta. This
variable defaults to false. The value of this variable is checked
during the validation phase for enterprise deployment types.
* If true, the install will not continue.
* If false, The user is presented with an informative message letting
them know this is beta software and there is low/no support at this
time. The installation will abort and instruct the user how to
continue.
|
|
|
|
Automatic merge from submit-queue.
Enable SC, ASB, TSB by default
It was decided that SC, ASB, and TSB should be installed by default.
This will not install the components on an upgrade from 3.6 to 3.7, that will require that you run the sc playbook on your own.
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1508085
|
|
Automatic merge from submit-queue.
openshift_hosted: Add docker-gc
Two new inventory variables have been created:
- ``openshift_crio_enable_docker_gc``: Enable docker_gc daemon set
- ``openshift_crio_docker_gc_node_selector``: Optional dictionary to use node
selector
When ``openshift_crio_enable_docker_gc`` and ``openshift_use_crio`` are both true
then ``docker_gc`` daemonset will be created along with adding a ``docker-gc``
sa.
|
|
Currently, upgrade_control_plane calls tasks meant for
master on etcd group.
This commit corrects the groupname.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1508734
|
|
Two new inventory variables have been created:
- openshift_crio_enable_docker_gc: Enable docker_gc daemon set
- openshift_crio_docker_gc_node_selector: Optional dictionary to use node
selector
When openshift_crio_enable_docker_gc and openshift_use_crio are both true
then dockergc daemonset will be created along with adding a dockergc
sa.
Signed-off-by: Steve Milner <smilner@redhat.com>
|
|
This will not install the components on an upgrade.
|
|
|
|
Automatic merge from submit-queue.
GCP cannot use growpart utils from AWS due to conflicting package
Allow cloud provider specific packages to be specified. Also, fix a wait condition in openshift_gcp to allow masters to be bootstrapped nodes.
@kwoodson
|
|
Automatic merge from submit-queue.
Create default storageclass for cloudprovider openstack
With this patch, a default storageclass will be created on openstack cloudprovider after installation.
|
|
Make the required package optional
|
|
Automatic merge from submit-queue.
Correctly install cockpit
Seems this was changed inadvertantly in #5378
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1507061
|
|
Add dm_thin_pool for gluster use
|
|
|
|
Seems this was changed inadvertantly in #5378
|
|
Force include_role to static for loading openshift_facts module
|
|
1504593 Refactor health check playbooks
|
|
This previously passed all CI tests, merge queue broke last night.
Merging, this is a blocker bug.
|
|
|
|
Moving removal of unwanted artifacts to image_prep.
|
|
Documentation fix for openshift_aws_iam_cert_key_path
|
|
|
|
Automatic merge from submit-queue.
Openshift MGMT Fixes and Container Provider Integration
Various small fixes/polishing to the openshift management role.
* Updated uninstall script
* Automatically add cluster to new management deployment
* Scaffolding for scripts to add *N* number of OCP/Origin clusters as Container Providers in any arbitrary MIQ/CFME instance
* Fix mis-named task in `openshift_nfs/create_export`
* Uninstall playbook only runs on one master
|
|
Automatic merge from submit-queue.
Ensure deployment_subtype is set within openshift_sanitize_inventory.
https://bugzilla.redhat.com/show_bug.cgi?id=1504021
|
|
Remove base package install
|
|
|
|
Automatic merge from submit-queue.
Add installer checkpoint for prometheus
|
|
Currently, in 1.5, 3.6, 1.7 upgrade-all plays, control
plane upgrades are not called correctly.
This commit ensures the master config hook is appropriately
applied during these upgrades to match the steps in
control plane only upgrades.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1486054
|
|
Automatic merge from submit-queue.
Enable oreg_auth credential replace during upgrades
Currently, upgrades run a docker image pull prior to
upgrading masters and nodes for containerized installs.
If using a secure registry, and a user wishes to upgrade
their credentials due to expiry, the image pull will fail.
This commit ensures docker login credentials are updated
during upgrades, if necessary.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1503995
|
|
ingvagabund/strip-dash-when-comparing-versions-python3
strip dash when comparing version with Python3
|
|
Automatic merge from submit-queue.
Ensure upgrades apply latest journald settings
Currently, existing clusters might not have journald
configurations applied. This may result in a rate-
limiting of important log messages on openshift-masters.
This commit ensures that journald settings are applied
during the upgrade process openshif-masters.
Fixes: https://github.com/openshift/openshift-ansible/issues/5642
|
|
Generate internal hostnames of no_proxy
|
|
Correct version gates on policy reconciliation and add retries
|
|
Check if the master service is non-ha or not
|
|
Currently, base atomic-openshift package is
installed for versioning.
This doesn't appear to be necessary.
This commit removes this step.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1504196
|
|
|
|
1504515 Correct host group for controller restart
|