It’s my first publication so excuse me if I skip any rule, we will progress along the way 😉

My problem:
I have an application container that in opportunities when a new revision comes on the way it stays in activating state.
When checking the logs, it only stops reporting at the time that a new revision started to be generated until one has to stop and start again the application container.
It is an API application, which is composed of nodejs, redis and a postgresql.
The first 2 components run in the same container and the postgresql is a dedicated resource.
The exposed port is 3000.
I filed a case with Microsoft and so far no response.
Another thing to note is that the solution was brought in by an external provider, my job is to manage the resources in Azure.
If you need any particular information, I will share it with you, I have a lot, so I don’t want to fill with evidence that may not be useful to the case.

The most recent changes I have made are to enable session affinity, which from what I understand will help direct traffic to new replicas if the previous one is not available and leave 2 replicas always active.
I use the maximum that I am allowed by the consumption plan that I use, which would be 2vCPU and 4 GiB RAM.