I have a requirement to toggle service model behavior at run-time. Is there any inbuilt feature to achieve it. The purpose of this use case is to reduce incoming load on the live service and implement caching mechanism within the CA DevTest service.
I am interested in 2 model behaviors -
1. Provide virtualized responses
2. Pass requests to the live system
My requirement is as below -
- When first client request comes to the service
- service will check in the data set for any previous request-response pair
- If not found then service will pass request to the the live back-end service
- Store request-response pair in data set for future use
- Send response back to the client
- When same 2nd client request comes to the service
- If found then service will respond back with the response from data set
You mean like healing mode?
Session Viewing and Model Healing - DevTest Solutions - 9.1 - CA Technologies Documentatio
Thanks Ben. It will be helpful but I want to automate switching of model behavior. The session viewing and model healing is related to the virtual service after the deployment.
You could use the Stand In mode where the. Retreat first goes to the VS and if monarch found it is routed to the Live System.
To do the second one - you'd use Learning Mode where since the response will come from the Live System, it will be added to the running system and the next time the same request comes in, the response is from the VS based on what was learnt.
Thanks Koustubh. Let me try out the option. My service type is http. As per my understanding I have to keep my default mode as Stand In and if request-response pair not found in in-memory data set then changed the mode to Learning. I have to toggle between these 2 modes.
Did Koustubh's recommendations help resolve this issue?