Smoke test error while scaling app instances "[FAILED] Expected to hit all 2 app instances in 120 attempts, but didn't", on TAS 6.0.5

Products

• VMware Tanzu Application Service

Issue/Introduction

In TAS v4.0.25 & v6.0.5, smoke-test fails due to the app reporting RUNNING state when its actually in STARTING state. Examples of this behavior illustrated below (when scaling an app):

 

Below is an example of the error returned when app instance is showing RUNNING and is expecting 200's status codes but its actually unreachable and returning 400's because the state is actually STARTING, resulting in the "Bad Request":

400 Bad Request: Requested instance ('1') with guid ('20b#####-6###-4##l-a###-5#########2d') does not exist for route

 

Environment

Affected Versions:

• TAS v4.0.25 & v6.0.5

Cause

This is a known issue with the CAPI version that TAS is using.

• There is an discrepancy with how CAPI is reporting the status of an app (starting vs running state) when scaling the said application.  

Resolution

The fix is included in capi 1.192.0, which is expected to be included with TAS v6.0.8.

The current workaround is to upgrade capi to v1.192.0.

 

Additional Information

github:

• https://github.com/cloudfoundry/cloud_controller_ng/commit/b8eaeda331be8b1b50c438fe685ec70ee784f8bb