Story (Required)

As a QE trying to automate manual tests, I want to make tasks inside a pipeline independent of previous task's result 

Background (Required)

With current implementation, a pipeline task failure results in subsequent tasks' failure causing them to be dependent. This means that even if tasks are independent technically, the second task execution would fail if the first task fails which is not desirable. 

Out of scope

<Defines what is not included in this story>

Approach (Required)

There is a concept "onError" in Tekton, using which we can continue a task even if  there are failures. We need to handle the failures in a separate task and evaluate which tasks have failed and finally mark the new task failed if any failures are found.  The disadvantage would be that a task will be marked green even if it fails but this can be vied as an item in the list in the last task

1. Update existing pipeline include "on Error" option pipelines/acceptance-tests-release.yaml 

2. Evaluate and handle failures in a new task.

Dependencies

<Describes what this story depends on. Dependent Stories and EPICs should be linked to the story.>

Acceptance Criteria (Mandatory)

• One task failure inside a Pipeline does not block the subsequent tasks from executing 
• Final task gives a clear idea on which all previous tasks have been failed
• Results from all the tasks are available if necessary 

INVEST Checklist

Dependencies identified

Blockers noted and expected delivery timelines set

Design is implementable

Acceptance criteria agreed upon

Story estimated

Legend

Unknown

Verified

Unsatisfied

Done Checklist

• Code is completed, reviewed and checked in
• Continuous Delivery pipeline(s) is able to proceed with new code included
• Acceptance criteria are met