Open evanlouie opened 5 years ago
@edaena: after talking a bit w/ @andrebriggs, we should probably shelve this for a bit until we scope out what Azure DevOps is able to do overall (and what it can facilitate for us in a workflow).
@andrebriggs This seems to be getting quite stale. Is this something we still want to look into?
For an ideal Canary deployment scenario, there should be 1+ Azure DevOps deployment gates which connects to the monitoring source related to the application health to determine whether or not to proceed with the next phase of deployment (i.e. to increase canary % of traffic or deploy from staging to prod).
To best accomplish this, a Prometheus deployment gate extension should be created which is similar to the Azure monitor deployment gate extension. This would query the Prometheus API.
Related articles: https://docs.microsoft.com/en-us/azure/azure-monitor/app/continuous-monitoring https://github.com/Microsoft/azure-pipelines-tasks/blob/master/docs/authoring/gates.md https://docs.microsoft.com/en-us/azure/devops/pipelines/tasks/utility/azure-monitor?view=azure-devops https://docs.microsoft.com/en-us/azure/devops/pipelines/release/approvals/index?view=azure-devops
https://github.com/CatalystCode/kubemalt/issues/45