Hello folks, I am trying to deploy 1.2.0 onto a small Kubernetes cluster and facing some issues with some of the pods deploying, and I was wondering if someone else perhaps faced the same issue. I've deployed 1.1.0 before and didn't face any difficulties before so I'm a bit puzzled as to what might've changed. The thing I'm stuck on currently is the following:
I deployed onto a cluster using the Helm chart (under bin/k8s) and most everything went smoothly; in fact the UI comes up and works to an extent. However the CU manager is stuck in initializing, because the lakekeeper-init pod didn't complete successfully. I looked at the logs and there doesn't seem to be any useful output from this failed pod. All I can see in the logs is that it successfully got to at least the part of its script where it installs some Alpine packages, but there's no output past that. Presumably this means something went wrong before the line that echoes "Waiting for Lakekeeper health endpoint...", but it's hard to imagine what exactly that could be. The obvious culprit would be the wget for the mc client but I would imagine that would send something to stderr if it did fail. I think my next step is to alter this pod's command to use sh in debug mode to see where precisely it's failing, but I thought I would ask here in case someone already experienced this issue and has the resolution at hand. Thanks, - Ian
