| Age | Commit message (Collapse) | Author |
|---|
|
Clarify requirement of having etcd group
|
|
Use openshift.common.ip rather than ansible_default_ipv4 in etcd migration playbook.
|
|
Bug 1490304: Etcd scale-up playbook should add new member to etcdClientInfo of master-config.yaml
|
|
Don't assume storage_migration control variables are already boolean
|
|
playbook.
|
|
Merged by openshift-bot
|
|
|
|
openshift_upgrade_pre_storage_migration_enabled
openshift_upgrade_pre_storage_migration_fatal
openshift_upgrade_post_storage_migration_enabled
openshift_upgrade_post_storage_migration_enabled
because the 4 variables are not already defaulted/evaluated there's no
guarantee that they're actually the boolean type at this point.
Example: if they're passed in on the command line
|
|
Signed-off-by: Steve Milner <smilner@redhat.com>
|
|
|
|
Recent refactoring to remove openshift_common resulted
in base openshift rpm's being installed on more hosts
than previous. This situation results in hosts that
would otherwise not need access to openshift repositories
to require them.
This patch set results in only openshift_masters and
openshift_nodes to have the openshift base package installed.
|
|
|
|
Add openshift_public_hostname length check
|
|
Revert back to hostnamectl and previous default of not setting hostname
|
|
Merged by openshift-bot
|
|
upgrade: Verify required network items are set
|
|
|
|
Correct include path to not follow symlink
|
|
Merged by openshift-bot
|
|
|
|
Currently, the variable openshift_hostname is checked to ensure it
is less than 63 characters.
The variable openshift_public_hostname should also be checked for
this length.
This commit checks the length of openshift_public_hostname to ensure
it is 63 characters or less.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1467790
|
|
Currently, the variable openshift_pkg_version is defined
by the first master unless specified by the user. The role
openshift_version attempts to set this value on all other
hosts after setting it on the first master.
During normal installation, openshift_pkg_version is not
initially defined, therefor each host will utilize the
most recent rpm available. This creates a problem
during later operations of a cluster if a node is added
or replaced by a new node.
This commit ensures that the value for openshift_pkg_version
set by the first master is respected by subsequent hosts.
Fixes: #5341
|
|
The old default was "{{ not openshift.common.version_gte_3_1_or_1_1 }}"
just default to false instead, master branch will never be used with 3.0
hostname module doesn't work on atomic host
Fixes BZ 1489913
|
|
|
|
|
|
When upgrading osm_cluster_network_cidr, osm_host_subnet_length, and
openshift_portal_net must be set to avoid SDN initialization errors.
This was found when the default parameters were changed between Openshift
versions. This meant users who upgraded and did not specify either
mentioned variable at install/upgrade time ended up getting SDN errors
post upgrade.
When osm_cluster_network_cidr, osm_host_subnet_length, and
openshift_portal_net are not set the upgrade will fail telling the user
that the variables must be set and how to find the current values in the
current install.
References: https://github.com/openshift/openshift-ansible/commit/b50b4ea0b03feb9431abd7294fe4fb6b549ddfc0
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1451023
|
|
|
|
Provisioning updates.
|
|
|
|
Merged by openshift-bot
|
|
Merged by openshift-bot
|
|
Most of this role's purpose was to set facts. The vast majority
of these facts were simply redefining user-supplied variables.
This commit also removes various artifacts leftover from
previous versions, as well as variables that seem to be
entirely unused.
|
|
A new role for installing prometheus on openshift.
Depends on `openshift_hosted_prometheus_deploy` flag
role creates:
- prometheus namespace
- prometheus clusterrolebinding and service account
- pvs for prometheus, alertmanager and alertbuffer for internal nfs
- prometheus pod with prometheus behind oauth-proxy, alertmanager and alert-buffer behind oauth-proxy
- prometheus and alertmanager configmaps
- prometheus and alerts services and direct routes
- prometheus, alertmanager and alert-buffer pvcs
|
|
|
|
|
|
Added byo/openshift-master/config.yml as a new entry point for
performing master installation.
Excluder disable/enable for master has been moved to
common/openshift-master/config.yml to follow a standard pattern that
could be used for both installation and scaleup playbooks.
|
|
Merged by openshift-bot
|
|
Merged by openshift-bot
|
|
[BZ1486450] Only run migrate auth for < 3.7
|
|
|
|
Added byo/openshift-node/config.yml as a new entry point for performing
node installation. This playbook must be run after masters have already
been deployed.
Excluder disable/enable for both nodes and masters has been moved their
respective config.yml. This created a common pattern that could be used
for both installation and scaleup playbooks.
|
|
Merged by openshift-bot
|
|
Test: Fail on entry point playbooks in common
|
|
Merged by openshift-bot
|
|
Merged by openshift-bot
|
|
Move firewall install and fix scaleup playbooks
|
|
Fix scaleup on containerized installations
|
|
|
|
Currently, openshift_master upgrade play imports tasks directly
from the openshift_master role. This method does not honor
role defaults.
This commit changes sets openshift_master_config_dir if undefined.
|
|
|
