SimonEismann
commented
2 years ago Hi Sören, we previously ran the TeaStore with the Istio service mesh, so in theory, this should be possible.
In general, this sounds like there might be some networking issues where pods can not communicate with the persistence or the persistence can not connect to the database container. My initial theory would be that the persistence can not reach the database as this would cause the recommender/imageprovider to stay in the generating/waiting state and the persistence to be unreachable. Since you are getting the generating/waiting state, this means that the pods can communicate with the registry so there seems to be some network connectivity.
You are already on the right track, that the next step would be to check the logs of the persistence container. Here, it's quite odd that only the tomcat logs show up. Could you post an example of what the persistence logs currently look like?
SoerenHenning
Hey folks,
we just started playing around with TeaStore and are trying to run it along with Open Service Mesh (OSM) in Kubernetes. While without OSM everything's working fine, we experience some connection errors with sidecars injected by OSM. For example, the status page indicates that some services could not be reached or service remain in a generating/waiting state forever. We are now trying to investigate this further, but it's a bit difficult to find the root cause.
Is TeaStore generating any log messages? Especially while setting up the persistence some information about errors would be helpful. If TeaStore does, where can we find those logs?
$ kubectl logs -f teastore-persistence-... -c teastore-persistenceonly reports the default Tomcat logs.