Make identity hub to persistence

[LinxZW]

Background

I attempted to store additional Verifiable Credentials (VCs) in the Identity Hub (IH) of a consumer. The storage process works successfully, but when I stop and restart the container, the added credentials are removed. Only the ProcessorCredential and MembershipCredential remain.

Questions

1. According to the documentation, when using Kubernetes, data persistence is supported via PostgreSQL. However, this does not seem to work in my case. Am I misunderstanding something, or is there a configuration I might have missed?
2. To ensure data persistence for VCs stored in the IH, what steps or configurations should I implement?

[paullatzelsperger]

there is not a lot of information here. Assuming you are using the Verifiable Credentials API to store credentials (POST /v1alpha/participants/{participantContextId}/credentials), then the credential should indeed get stored in the database.

You can check the database, the table you're looking for is called credential_resource

[paullatzelsperger]

the URL you posted looks like the cluster-internal service name that is not accessible outside the cluster. The easiest way is to configure a port-forward for the postgres pod and inspect the DB thus.

[LinxZW]

I just gained access to the database (postgresql://consumer-postgres-service:5432/consumer, username: consumer, password: consumer). The credentials_resource table correctly shows the new credentials after creation:

However, after restarting the container, using Docker desktop

, the credentials disappear from the table:

Questions

1. Why are the credentials not persisting in the credentials_resource table after a container restart?
2. Could this be related to the database configuration or container setup (e.g., volumes, initialization)?
3. How can I troubleshoot and resolve this persistence issue?

[paullatzelsperger]

if you restarted the mvd-control-plane container, you didn't just restart the EDC control plane, instead what you did is restart the entire cluster.
KinD runs an entire K8S cluster inside a single Docker container (KinD = Kubernetes in Docker). So when you kill and restart the mvd-control-plane container, that actually restarts the entire cluster, not just a single connector's control plane.

The confusing bit is that a Kubernetes cluster also has a "control plane", which is something different entirely from the EDC control plane.

So what you did presumably, is kill and restart the cluster, which kills (deletes) all pods including their persistence.

I recommend using a tool like K9s or OpenLens to inspect the actual cluster.

[paullatzelsperger]

@LinxZW if this solves your problem, then please mark the answer