how are you running it?

Do you mean how am I running the query or how am I running the authz deployment? for the query, we checked the value in the db directly in the alembic_version table. For the authz deployment, we deployed in k8s with this config:

apiVersion: authzed.com/v1alpha1
kind: SpiceDBCluster
metadata:
  name: authz-ps
  namespace: dev
spec:
  config:
    datastoreEngine: postgres
  secretName: ...
  patches:
    - kind: Deployment
      patch:
        spec:
          template:
            spec:
              nodeSelector:
                ...

how are you running the migrate command

We are not running the migrate command specifically. When we are deploying the authz to k8s it's trying to do it automatically.

i.e. you're using the operator?

I'm sorry. I really don't understand what you mean by that. Here's the deployment yaml for the operator, if that helps. If not could you please tell me what I should look for? https://cdn.discordapp.com/attachments/1420792392793067681/1421139712655102012/message.txt?ex=68d7f2fe&is=68d6a17e&hm=cbd58612797eeaf084f18b7407ada570ea5fcd092f31f632f3deebc3d08a226f&

that looks like the deployment of the operator itself rather than the custom resource that you wrote to have the operator do the work. do you have a SpiceDBCluster [like this one](https://github.com/authzed/spicedb-operator?tab=readme-ov-file#suggested-updates) written somewhere? does that object have a

channel

defined on it?

Hi @yetitwo ,Thanks for the response. This is what our spicedb operator resource yaml looks like: https://cdn.discordapp.com/attachments/1420792392793067681/1422305165037404321/message.txt?ex=68dc3068&is=68dadee8&hm=fb35b074fb42106b79b8156d06139e4077de08ab6b0b75bb2c2eaf5a0b0b638a&

ah, i don't mean the definition of the operator

you mean this?

apiVersion: authzed.com/v1alpha1
kind: SpiceDBCluster
metadata:
  name: authz-ps
  namespace: dev
spec:
  config:
    datastoreEngine: postgres
  secretName: ...
  patches:
    - kind: Deployment
      patch:
        spec:
          template:
            spec:
              nodeSelector:
                ...

that I initially sent. All I can tell is that we definitely haven't defined any update channels

yeah, that was the part i wanted to pull at

Thanks @yetitwo We are now getting this error:

I0930 15:32:27.239033       1 event.go:389] "Event occurred" object="dev/devspice-new" fieldPath="" kind="SpiceDBCluster" apiVersion="authzed.com/v1alpha1" type="Warning" reason="InvalidSpiceDBConfig" message="invalid config: no update found in channel"

We tried these yamls:

apiVersion: authzed.com/v1alpha1
kind: SpiceDBCluster
metadata:
  name: devspice-new
  namespace: dev
spec:
  config:
    datastoreEngine: postgres
  secretName: newsecret

apiVersion: authzed.com/v1alpha1
kind: SpiceDBCluster
metadata:
  name: devspice-new
  namespace: dev
spec:
  channel: stable 
  version: v1.45.4
  config:
    datastoreEngine: postgres
  secretName: newsecret

apiVersion: authzed.com/v1alpha1
kind: SpiceDBCluster
metadata:
  name: devspice-new
  namespace: dev
spec:
  channel: stable 
  version: v1.45.4
  config:
    datastoreEngine: postgres
  secretName: newsecret
status:
  version:
    name: v1.14.0
    channel: stable 
  availableVersions:
  - name: v1.14.1
    channel: stable
    description: direct update with no migrations

v1.45.4

isn't available in

stable

yet - the most recent version is available in rapid, and the second-to-most-recent is available in

stable

Now it works. Thank you so much @yetitwo

sure thing!