| Age | Commit message (Collapse) | Author |
|
Automatic merge from submit-queue.
Bug 1535947: Fix missing task in metrics, logging uninstall playbooks
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1535947
/assign @sdodson
|
|
Automatic merge from submit-queue.
Ensure atomic_proxies are configured with docker
In some setups, system containers may be used
in conjunction with package_docker.
This commit ensures that the atomic
proxies are configured while using package_docker
and containerized components.
This commit also ensures containerized_groups
are based on openshift_is_containerized instead
of 'containerized' variable directly to
account for possible use of atomic hosts which
users may not specify containerized in inventory
directly.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1535323
|
|
Automatic merge from submit-queue.
Fix flake8 errors in utils/test
|
|
|
|
|
|
Make openshift_web_console_prefix defaults like other components
|
|
|
|
Currently, users have no way to run preqrequisites.yml
on just newly added nodes during scaleup.
This commit ensures only the new nodes are changed during
scaleup as well as ensure prerequisites are run.
|
|
The console will discover if the broker is running, but only on startup.
Trigger a rollout so that the console sees the change.
|
|
Automatic merge from submit-queue.
Bug 1509082 - Adding auth as a secret
If username and password are defined we should use a secret for the credentials.
|
|
Automatic merge from submit-queue.
Add Azure to support openshift_cloudprovider_kind
Try to fix this as an example. I checked it works.
|
|
This commit adds the ability to mount volumes into
system containerized nodes.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534933
|
|
update health check required versions for 3.9
|
|
installing etcd
This allows non-Atomic hosts to be used to install containerized etcd
|
|
Automatic merge from submit-queue.
kibana checks: use six.moves instead of ImportError
|
|
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1535947
|
|
Other components are like 'registry.access.redhat.com/ose-' or
'openshift/origin-' based on OCP versus Origin.
|
|
Right now this is only available on fedora so guard it with
openshift_enable_unsupported_configurations
|
|
This was introduced as a temporary fix for the origin+centos
deployments. It is no longer necessary and it's confusing to
newcomers.
|
|
Automatic merge from submit-queue.
Rework default TSB prefix and imagename to match other services
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1518917
|
|
Automatic merge from submit-queue.
Ensure we are running oc execs against running pods
We should make sure we are running `oc exec` statements against running pods only.
Addresses https://bugzilla.redhat.com/show_bug.cgi?id=1534879
|
|
Automatic merge from submit-queue.
Setting default storage_class_names for when calling openshift_loggin…
…g_elasticsearch role
Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=1521218
|
|
|
|
Bug 1511357: Task "Reload generated facts" failed while scaling up containerized masters
|
|
Add new grafana playbook.
|
|
|
|
|
|
Created by command:
/usr/bin/tito tag --debug --accept-auto-changelog --keep-version --debug
|
|
openshift_checks: repair adhoc list-checks mode
|
|
Automatic merge from submit-queue.
Check that rc != 0 for steps with 'failed_when'
Some parts don't check the result rc at all, some check that 'error' is present there.
Instead the safest way is to check for both
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1534538
|
|
Add init/main.yml to etc-upgrade
|
|
The prometheus block duration settings should not be changed from their defaults.
See also: https://github.com/prometheus/prometheus/pull/3618
|
|
juanvallejo/jvallejo/fix-elasticsearch-diskspace-check
Automatic merge from submit-queue.
include elasticsearch container name
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1534988
Addresses cases where the elasticsearch pod with multiple containers does not have the `elasticsearch` container as its first one.
cc @sosiouxme
|
|
|
|
In some setups, system containers may be used
in conjunction with package_docker.
This commit ensures that the atomic
proxies are configured while using package_docker
and containerized components.
This commit also ensures containerized_groups
are based on openshift_is_containerized instead
of 'containerized' variable directly to
account for possible use of atomic hosts which
users may not specify containerized in inventory
directly.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1535323
|
|
|
|
Faciliate 3.7 to 3.8 to 3.9 upgrade
|
|
|
|
|
|
There are some obsoletes in 3.9 packages which lead yum to upgrade
directly from 3.7 to 3.9 when you ask for 3.8 packages while 3.9 repos
are enabled. Since we'd like to allow people to run one playbook to
upgrade from 3.7 to 3.8 to 3.9 we need to exclude those packages when
upgrading to 3.8
|
|
|
|
|
|
Currently, etcd upgrade playbook only calls
evaluate_groups to limit the number of tasks
that are run against nodes during initialization.
Recently, changes have been added to allow skipping
of most node tasks via providing variablized host
groups during initialization code.
This commit allows etcd upgrades to use common
init code to ensure necessary facts are defined
and proper variables are sanity checked.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1527771
|
|
|
|
|
|
|
|
Don't use from ansible.module_utils.six as its no longer available in Ansible 2.4
|
|
Add missing task that got dropped in a refactor
|
|
Automatic merge from submit-queue.
Move OpenStack inventory.py out of sample-inventory
This moves the dynamic inventory for OpenStack from the `sample-inventory` dir users are expected to copy and edit to `/inventory/openstack`.
This will ensure that people don't use an outdated inventory as well as communicates that the inventory (similar to the playbooks and roles) is not something they are expected to modify under normal circumstances.
The sample inventory will now contain only `group_vars` and as such, moving from that to providing vars explicitly is also more obvious (you replace the `-i inventory` command with `-e @vars.yaml` and you're good to go).
|
|
|