failed container has no logs

[shanproofpoint]

Hi,

I am having a lot of trouble trying to find out why a container fails because kubedock prunes it too quickly. Also no logs come through to the testcontainers-java side even after

container.withLogConsumer(l-> System.out.println(serviceName + ":" + l.getUtf8StringWithoutLineEnding()));

i could not find a configuration to keep the pod around per docker request to create a container.

also is it possible to setup a different pod template per container being requested as well instead of a global one using --pod-template ?

I also cannot find an example of how a pod template is parameterized... which sections will you overwrite from the template?

[joyrex2001]

I am having a lot of trouble trying to find out why a container fails because kubedock prunes it too quickly. Also no logs come through to the testcontainers-java side even after

I don't think I will be able to help you with that; if the pod needs to be there for a bit longer, I think the best approach would be to somehow make it pause in the test.

is it possible to setup a different pod template per container being requested as well instead of a global one using --pod-template ?

At the moment this is not possible, the pod-template is a generic config currently. What do you think would be a good way to influence that behaviour from your point of view?

I also cannot find an example of how a pod template is parameterized... which sections will you overwrite from the template?

In general, when the pod is created, it will first load the pod-template, and after that it will add the containers and additional volumes (e.g. for optional init and sidecar containers). If a container is added in the pod-template, each created container will be based on that config.