Continuous Delivery with Docker and Jenkins

Rolling back deployments

I remember the words of my colleague, a senior architect, You don't need more QAs, you need a faster rollback. While this statement is oversimplified and the QA team is often of great value, there is a lot of truth in this sentence. Think about it; if you introduce a bug in the production but roll it back soon after the first user reports an error, then usually nothing bad happens. On the other hand, if production errors are rare but no rollback is applied, then the process to debug the production usually ends up in long sleepless nights and a number of dissatisfied users. This is why we need to think about the rollback strategy up front while creating the Jenkins pipeline.

In the context of Continuous Delivery, there ...