Add resources required to run lightspeed-tests - #4140
Conversation
Updated the playbook to work with the current OpenStack Lightspeed operator after org migration from openstack-lightspeed to openstack-k8s-operators. Changes: 1. Namespace: Changed default from openshift-lightspeed to openstack-lightspeed to match operator's suggested namespace in csv 2. CR creation: Added STEP 3 to create OpenStackLightspeed CR - Deploys the actual service (not just operator) - Configurable LLM parameters (endpoint, credentials, model) - Graceful timeout with ignore_errors for test environments 3. Kubeconfig fix: Resolved variable recursion bug on line 29 - Renamed cifmw_openshift_kubeconfig to kubeconfig_path - Prevents 'recursive loop detected' error Relates-To: OSPRH-33342" Signed-off-by: malingatembo <tembomalinga@gmail.com>
Signed-off-by: malingatembo <tembomalinga@gmail.com>
Fixes the following 3 blocking issues 1. Remove hardcoded Downstream LLM configurations - Make llmEndpoint and modelName required variables - Update README to document these vars as REQUIRED 2. Add llm credentials secret creation - Create secret from cifmw_openstack_lightspeed_api_token when provided - Validate secret exists when variable not provided - Fail fast when secret missing ( with clear message ) 3. Add TLS certificate ConfigMap creation - Create ConfigMap from cifmw_openstack_lightspeed_ca_cert when provided - Validate ConfigMap exists when variable not provided - Fail early with clear error message if ConfigMap missing - Provides flexibility: either create via playbook or pre-create manually Signed-off-by: Malinga Tembo <mtembo@redhat.com>
this pr: 1. removes CR wait task (fixes ignore_errors issue) - ci job calling this hook should now be responsible for waiting for full service health 2. Add no_log to secret creation 3. Add CA and URL download support Signed-off-by: Malinga Tembo <mtembo@redhat.com>
The example incorrectly showed HTTP. Updated to HTTPS and documented the requirement. In fact RH cert server uses HTTPS. Fixes coderabbit security concern. Signed-off-by: Malinga Tembo <mtembo@redhat.com>"
Per lpiwowar feedback: - Add no_log to secret verification task (prevent token leak with -vvvv) - Merge two ConfigMap creation tasks into one with conditional logic Signed-off-by: Malinga Tembo <mtembo@redhat.com>
Qodo reviews are paused for this user.Troubleshooting steps vary by plan Learn more → On a Teams plan? Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center? |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
test-operator expects a ConfigMap and two secrets to exists before it will start executing a test suite. We create the expected dummy resources with not data to satisfy the requirement. test-operator is using the default ServiceAccount to run this test. We need to create a ClusterRole and ClusterRoleBinding to grant the default SA access to the Lightspeed service. Signed-off-by: Jan Cervenka <jcervenk@redhat.com>
77088b0 to
9c76c36
Compare
|
Hi @malingatembo! we were discussing with @lpiwowar (see: openstack-k8s-operators/lightspeed-tests#12) create some additional resources in your playbook. The resources are required by the test-operator to successfully run the lightspeed-test suite. See the commit message:
|
|
@jancervenka thanks for the PR! 🎉 The initial idea was for this PR to build on top of @malingatembo's PR. We can then use If it turns out that the changes you introduced work, then we can merge both PRs into one (into @malingatembo's PR #4108 with a proper commit attributing your work, of course :). I suggest:
Just want to say that we are getting a little bit ahead ourselves here. But if everything clicks then the PR merging is going to be easy. Also, I'm personally ok just waiting it out until the job @malingatembo is working on gets ready. One way or the other I would mark this as a draft for now 🙈 . |
Qodo reviews are paused for this user.Troubleshooting steps vary by plan Learn more → On a Teams plan? Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center? |
|
the PR is checked out of @malingatembo's work and marked as a draft. |
3 participants
test-operator expects a ConfigMap and two secrets to exists before it will start executing a test suite. We create the expected dummy resources with no data to satisfy the requirement.
test-operator is using the default ServiceAccount to run the tests. We need to create a ClusterRole and ClusterRoleBinding to grant the default SA access to the Lightspeed service.