How to see the status and access logs

Check the status of the pods:

kubectl get pods

...

Reading the logs from a pod
kubectl logs button3d-79447648c-s4x7x


Redirect the logs to a file on the server
kubectl logs button3d-79447648c-s4x7x > button3d.log

The Application is Not Reachable and returns 400 Bad Request

This means, that you are trying to reach the application through a hostname that has not been configured in the admin panel, or that you haven't loaded any organization data into the application.

Please make sure, that you are using the main dns configured for the application when accessing it, and not the IP of the application server.

The Application Returns 502 Bad Gateway

502 Bad Gateway Indicates, that the application is listening and the reverse proxy is running, but it can not connect to one of the downstream components.

Check with kubectl get pods if all of the pods are running.

The Application Responds With a 500 Server Error

This means, that something crucial is not working in the button3d pod. First order of business is to try to recreate the application stack by logging into the server and running:

1. Inspect the button3d logs to find a stacktrace.

kubectl logs --tail 3000 $(kubectl get pods | grep "button3d" | cut -d" " -f1)

2. Uninstall the application stack

./helm uninstall order-management

3. Install the application stack

./helm upgrade -i order-management -f configure/values-on-premise.yaml -f=values.yaml chart/order-management-*.tgz

3D File Upload Fails Consistently

First, make sure that the 3D-Backend is actually running. Run the following command on your own workstation to check whether or not the 3D-Backend responds to requests:

...

Then backend is running.

Next, check the logs of the backend pod.

ValueError on migrations

If you see something like this during the initial migration:

...

it might be that a previous migration was interrupted and the database has an invalid state now.
Try to delete and recreate the database. We only encountered this error so far on initial database migrations.

Metabase embedded dashboards don’t load

Embedded dashboards that fail to load could be caused by a changed administrator password for Metabase. If you changed the password for the main admin in the UI, you will have to also update the corresponding adminPass value in the Metabase section of values.yaml. Run ./upgrade.sh inside the package afterwards and try reloading the embedded dashboards.