| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
Global proxies were not supported until 3.2 so if a
lesser version is specified then we don't need to
ask for it.
|
|
|
|
|
| |
Some UI improvements to help make it more clear to the user what information we
are looking for.
|
| |
|
|\
| |
| | |
a-o-i: UI additions for proxies
|
| | |
|
| | |
|
| |
| |
| |
| | |
Fix issue with assigning the Host object to a variable that expects a string.
|
|/
|
|
|
|
|
| |
Always prompt the user for a storage host. The prompt defaults to the first master and warns the
user not to use a container-based installation for this.
WIP
|
|\
| |
| | |
Add --gen-inventory command to atomic-openshift-installer.
|
| |
| |
| |
| |
| |
| |
| |
| | |
This will just generate the Ansible inventory once all information is gathered,
inform the user where to find it, and exit.
Includes some test utility updates to make what broke as a result of this
change less fragile.
|
|\ \
| | |
| | | |
a-o-i: Error out early if callback_facts is None
|
| | |
| | |
| | |
| | |
| | |
| | | |
Bail out much earlier if for some reason callback_facts does not
return any data. This should prevent later methods from generating
stack traces.
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Adds the ability to perform upgrades without any user input through:
atomic-openshift-installer --unattended upgrade --next-major
or
atomic-openshift-installer --unattended upgrade --latest-minor
|
| |/
|/|
| |
| | |
Change the upgrade logic to allow for more than two types of upgrades.
|
| |
| |
| |
| | |
Fix the nosetests after the rebase
|
|/ |
|
| |
|
|
|
|
|
|
| |
The previous check was less than reliable thanks to
potential differences between hostname / ip / connect_to.
This check should cover all instances.
|
|
|
|
| |
Some additional error checking when selecting the master_lb
|
|
|
|
|
|
|
|
|
| |
After much research and debugging, it turns out the facts
for nativeha hosts were much less consistent than I hoped.
With this newfound knowledge, I've implemented this method.
In short, you check with the first master to see what it thinks
the master_lb is, then you include that host as "installed".
|
|\
| |
| | |
a-o-i: pylint fixes related to too-long lines
|
| | |
|
|/ |
|
| |
|
|
|
|
|
| |
Updates to the upgrade command to allow the user to select a minor
upgrade.
|
|
|
|
| |
Add a question to ask the user for a default subdomain.
|
|
|
|
|
| |
This removes the option to specify a containerized install when
installing 3.0 in interactive mode.
|
|
|
|
|
|
|
|
| |
Fix error handling for invalid hostnames. Previously we were trying to print
the offending hostname out, which caused errors due to python's handling
of unicode strings. Seeing as how the user's input stays on the screen
directly above the error there is no need to try and print it and we can
avoid this problem entirely.
|
|\
| |
| | |
atomic-openshift-installer: Populate new_nodes group
|
| |
| |
| |
| | |
Set the new_nodes group when scaling up additional nodes
|
|/ |
|
|
|
|
| |
This is a fix for the failing test_fresh_install_with_config.
|
|
|
|
|
| |
Small update to remove the hint for multiple masters if we are doing
a 3.0 installation.
|
| |
|
| |
|
|
|
|
|
|
| |
Displays each host and the roles it will play based on the current
configuration. As the configuration grows the summary will adapt to indicate
embedded vs separte etcd, scheduled vs unscheduled nodes, etc.
|
| |
|
|\ |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If only 2 masters are specified, consider this a configuration error if running
an unattended install, and prevent it completely if running an attended
install. (continues to prompt for hosts until you have at least 3)
Because this condition cannot be entered in the interactive install, we can't
really write a test for this negative case.
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
Checking behavior when there is no LB specified, and when the user attempts
to re-use a master or node as their LB.
|
| | |
|
|/
|
|
|
|
|
|
|
|
|
|
| |
Code was present to catch this in unattended installs but was looking for a
host record with both master/node and master_lb set to true, but in the
attended installs we were adding a separate host record with the same
connect_to.
Attended tests can now optionally specify multiple "attempted" strings for the
master_lb specification, we'll try to input each if multiple are specified.
Cleanup some empty defaults and error messages as well.
|