| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Also:
* minor rewording of the text that informs the admin about scheduleable
masters.
|
|
|
|
|
|
The preconfigured load balancers, previously denoted by having 'run_on' set to
false, cannot have their facts gathered which results in a stack trace. Later
when we write out the inventory we have to fake out the hostname and just use
'connect_to'.
We're likely going to have the concept of other types of "plug-in" hosts where
we don't run ansible. We should make sure we abstract this properly so it's
easy to add additional types of hosts.
Also in the commit:
- Renamed 'run_on' to 'preconfigured' and inverted the logic as needed
- Output tally of Masters and Nodes as well as remaining Masters required for
HA
- Minor rewording in a few places
- Currently only prompting for the load balancer after all other hosts have
been entered
- Removed spurious echo
|
|
Rename ha_proxy variables and methods to 'master_lb' to better
future-proof things.
|
|
Reverse the order we ask two questions: What variant the user wants
to install and which hosts to install on. This lets us avoid asking
for multiple masters for 3.0 installs.
|
|
|
|
|
|
|
|
|
|
This adds the ability to quickly set up a multi-master environment.
|
|
atomic-openshift-installer: connect_to error handling
|
|
|
|
|
|
Catch the exception that happens when connect_to isn't specified in installer.cfg.yaml
|
|
A few fixes to keep pylint happy.
|
|
Default to OSE 3.1 in interactive installer instead of 3.0.
|
|
Correct the case where the first host entered is not a master.
|
|
For now, we should restrict the quick installer to a single master.
This should change in the near future.
|
|
|
|
|
|
|
|
|
|
atomic-openshift-installer: Update prompts and help messages
|
|
This is an update to the no password sudo test meant to address some
weirdness around group vs. user permissions.
|
|
|
|
|
|
- Split playbooks into two, one for 3.0 minor upgrades and one for 3.0 to 3.1
upgrades
- Move upgrade playbooks to common/openshift/cluster/upgrades from adhoc
- Added a byo wrapper playbooks to set the groups based on the byo
conventions, other providers will need similar playbooks added eventually
- installer wrapper updates for refactored upgrade playbooks
- call new 3.0 to 3.1 upgrade playbook
- various fixes for edge cases I hit with a really old config laying
around.
- fix output of host facts to show connect_to value.
|
|
atomic-openshift-installer: Remove question for container install
|
|
|
|
Removing the option for a container-based install from the quick
installer with it is in tech preview.
|
|
If this file exists on disk, the installer will use it if the user didn't
specify an ansible config file on the CLI.
Rename share directory to match the rpm name. (utils vs util)
|
|
Test fixes related to connect_to
|
|
This generates the ansible inventory based on the pruned list of non-installed
hosts we've created rather than the full host list provided in installer.cfg.yaml
|
|
There the tests didn't know anything about connect_to and we had a case where
we weren't handling the migration from the 3.0 installer config format to 3.1
|
|
Updating the atomic-openshift-isntaller local connection logic for th…
|
|
Upgrade enhancements
|
|
connect_to addition.
|
|
Changes to installer.cfg.yaml to allow for better defaults in unattended mode.
Update example in the docs.
|
|
|
|
|
|
- Push config dir logic out of module and use host variables instead.
- Backup master config with ansible utility.
- Add error handling for the upgrade config module.
- Add verbose option to installer.
- Return details on what we changed when upgrading config.
- Cleanup use of first master.
- Don't install upgrade rpms to check what version we'll upgrade to.
|
|
|
|
TODO: We desparately need tests cases for:
- interactive with no config file
- interactive with config file and all installed hosts
- interactive with config file and no installed hosts
- interactive with config file and some installed some uninstalled hosts
- unattended with config file and all installed hosts (with and without --force)
- unattended with config file and no installed hosts (with and without --force)
- unattended with config file and some installed some uninstalled hosts (with and without --force)
|
|
Previously the output was a little confusing. We didn't display anything about
the uninstalled hosts.
|
|
all cases
|
|
|
|
Previously we were writing out a inventory like this:
~~~
[OSEv3:children]
masters
nodes
[OSEv3:vars]
ansible_ssh_user=root
deployment_type=openshift-enterprise
ansible_connection=local
[masters]
ose3-master.example.com openshift_hostname=ose3-master.example.com
[nodes]
ose3-master.example.com openshift_hostname=ose3-master.example.com
ose3-node1.example.com openshift_hostname=ose3-node1.example.com
ose3-node2.example.com openshift_hostname=ose3-node2.example.com
~~~
The problem with that is now all the hosts are consider local connections. In
addition our sudo check wasn't working as expected. We would check that we
have sudo, but the playbooks were not running with root privileges. When
gathering facts you'd hit:
~~~
__main__.OpenShiftFactsFileWriteError: Could not create fact file: /etc/ansible/facts.d/openshift.fact, error: [Errno 13] Permission denied: '/etc/ansible/facts.d/openshift.fact'
~~~
Instead the test for locale connections needs to be per host. Anytime we're not running as root we need `ansible_become` set:
~~~
ose3-master.example.com openshift_hostname=ose3-master.example.com ansible_connection=local ansible_become=true
~~~
|
|
https://bugzilla.redhat.com/show_bug.cgi?id=1274201#c13
|
