summaryrefslogtreecommitdiffstats
path: root/roles/openshift_health_checker/openshift_checks
Commit message (Collapse)AuthorAgeFilesLines
...
* | remove skopeo dependency on docker-pyjuanvallejo2017-05-191-78/+70
| |
* | improve error handling for missing varsjuanvallejo2017-05-191-48/+56
|/
* Merge pull request #3630 from juanvallejo/jvallejo/add-etcd-volume-checkOpenShift Bot2017-05-191-0/+58
|\ | | | | Merged by openshift-bot
| * revert role-specific var namejuanvallejo2017-05-151-1/+1
| |
| * Merge branch 'jvallejo/add-etcd-volume-check' of ↵juanvallejo2017-05-121-5/+7
| |\ | | | | | | | | | github.com:juanvallejo/openshift-ansible into jvallejo/add-etcd-volume-check
| | * Update variable name to standardRodolfo Carvalho2017-05-111-1/+1
| | | | | | | | | It was agreed to name role variables as `r_ROLE_NAME_VARIABLE_NAME`. Giving it a try.
| | * Make class attribute name shorterRodolfo Carvalho2017-05-111-4/+4
| | |
| | * Add module docstringRodolfo Carvalho2017-05-111-0/+2
| | |
| * | check if hostname is in list of etcd hostsjuanvallejo2017-05-121-3/+4
| |/
| * Update checkRodolfo Carvalho2017-05-101-35/+33
| |
| * int -> floatRodolfo Carvalho2017-05-101-3/+3
| | | | | | We don't need to convert to int and then to float. Read it as float from the start.
| * Remove vim lineRodolfo Carvalho2017-05-101-1/+0
| | | | | | It has been agreed that we don't use it any longer.
| * add etcd volume checkjuanvallejo2017-05-091-0/+58
| |
* | add etcd cluster size checkjuanvallejo2017-05-171-0/+84
| |
* | Simplify memory availability check, review testsRodolfo Carvalho2017-04-171-22/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | - Fix required memory for etcd hosts (10 -> 20 GB), as per documentation. - Some changes to make the code more similar to the similar DiskAvailability check. - Do not raise exception for hosts that do not have a recommended memory value (those are ignored anyway through `is_active`, so that was essentially dead code). - Test that the required memory is the max of the recommended memories for all groups assigned to a host. E.g. if a host is master and node, we should check that it has enough memory to be a master, because the memory requirement for a master is higher than for a node.
* | Simplify mixin classRodolfo Carvalho2017-04-171-10/+7
| | | | | | | | | | - Expose only is_active and no other method. - Move general comment to module docstring.
* | Simplify disk availability check, review testsRodolfo Carvalho2017-04-171-46/+38
| | | | | | | | | | | | | | | | | | | | | | - only support a fixed list of recommended values for now, no overwriting via Ansible variables (keep it simple, add features as needed). - implement is_active: run this check only for hosts that have a recommended disk space. - test priority of mount paths / and /var.
* | add disk and memory availability check testsjuanvallejo2017-04-171-1/+4
| |
* | add ram and storage preflight checkjuanvallejo2017-04-172-0/+108
|/
* Merge pull request #3887 from rhcarvalho/pkg-version-only-master-or-nodeOpenShift Bot2017-04-131-0/+7
|\ | | | | Merged by openshift-bot
| * Do not check package version on non-master/nodeRodolfo Carvalho2017-04-101-0/+7
| |
* | Merge pull request #3757 from ↵OpenShift Bot2017-04-121-0/+4
|\ \ | | | | | | | | | | | | juanvallejo/jvallejo/skip-yum-update-check-on-fedora Merged by openshift-bot
| * | skip PackageAvailability check if not yumjuanvallejo2017-04-111-0/+4
| |/
* / preflight checks: refactor and fix aos_versionLuke Meyer2017-04-081-5/+3
|/ | | | | | | | | | | Bring in openshift_repos to provide Origin repo before checks on Origin. For OCP we want the check to fail if both version 3.3 and version 3.4 are available - they shouldn't have both channels enabled. For Origin everything is in one repo so it's not surprising to find 1.4 and 1.5 versions available. Added unit tests as well.
* Add test scaffold for docker_image_availability.pyRodolfo Carvalho2017-03-281-0/+11
| | | | | The intention is to set a starting point and let another team member work on the code to gain experience with tests.
* Test OpenShift health check loaderRodolfo Carvalho2017-03-281-12/+15
|
* Rename module_executor -> execute_moduleRodolfo Carvalho2017-03-284-5/+10
| | | | | | | | It is a function/callable, the name should imply action, should be a verb and not a noun. Keep supporting the old name while we have PRs in-flight that use the old name.
* add docker_image_availability checkjuanvallejo2017-03-231-0/+168
| | | | | | | | This patch adds a check to ensure that required docker images are available in at least one of the registries supplied in an installation host. Images are available if they are either already present locally, or able to be inspected using Skopeo on one of the configured registries.
* Do not require python-six via openshift_factsRodolfo Carvalho2017-03-201-6/+3
| | | | | | | | | | | | | It turns out the code in roles/openshift_health_checker/openshift_checks/__init__.py had a mistake in the import statement, and not the case that `add_metaclass` is not available in the embedded `six` from `ansible.module_utils`. The fix is to correct the import and do not require python-six to be installed on target hosts. Note: now that we have some unit tests in the role, errors like that would have been caught.
* Add unit tests for existing health checksRodolfo Carvalho2017-03-171-1/+2
|
* Revert module_utils six for openshift_health_checkerJason DeTiberus2017-03-131-1/+3
| | | | | Embedded version of six in Ansible 2.2.1 does not support add_metaclass, so revert to using python-six.
* Use ansible.compat.six where possibleJason DeTiberus2017-03-091-2/+5
|
* Use six from ansible.module_utils for remote hostsJason DeTiberus2017-03-091-3/+2
|
* Renamed NotContainerized to NotContainerizedMixin and dropped no-memberSteve Milner2017-02-225-12/+10
|
* Do not hard code package namesRodolfo Carvalho2017-02-101-8/+5
|
* Refactor code to access values from task_varsRodolfo Carvalho2017-02-104-17/+25
|
* Introduce tag notation for checksRodolfo Carvalho2017-02-104-0/+12
| | | | This allows us to refer to a group of checks using a single handle.
* Replace multi-role checks with action pluginRodolfo Carvalho2017-02-105-0/+188
This approach should make it easier to add new checks without having to write lots of YAML and doing things against Ansible (e.g. ignore_errors). A single action plugin determines what checks to run per each host, including arguments to the check. A check is implemented as a class with a run method, with the same signature as an action plugin and module, and is normally backed by a regular Ansible module. Each check is implemented as a separate Python file. This allows whoever adds a new check to focus solely in a single Python module, and potentially an Ansible module within library/ too. All checks are automatically loaded, and only active checks that are requested by the playbook get executed.
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-05 23:36:33 +0000