Context:

As a CRC tenant, we make use of Pipeline Runs in order to deploy the new versions of our code to stage and production environments. It's relatively common to see an error like:

[2023-07-21 07:34:20] [ERROR] [saasherder.py:_check_image:1084] - [ccx-data-pipeline-clowder/ccx-insights-results] https://github.com/RedHatInsights/insights-results-aggregator/blob/master/deploy/clowdapp.yaml: Image does not exist: quay.io/cloudservices/insights-results-aggregator:8d012ef 

which means that the deploy pipeline run before the image was built. I've also seen this issue for other teams deployments like "host-inventory".

This is easily solved by rerunning the pipeline, which is automated in most cases I would say. But what if the pipeline is not rerun? We would miss that deployment for stage/prod.

Some more context in the initial Slack discussion.

Useful links:

In our specific case we are using the insights-gh-build-master template, and I guess that most of the CRC should be using this one too or the GitLab alternatives. So, I think this is an issue affecting most of the teams in CRC.

Acceptance criteria:

The pipelines are not run until the image is pushed to quay.