In Helm commands, what is the cluster ID?

[nikoChen-I755695]

According to the instructions in Step 4 of the SetUp section in README.md, we successfully migrated a resource created by BTPO. However, when dealing with a manually created resource, we were unable to retrieve its cluster ID.
Based on the commands in this document, we couldn’t find the corresponding ConfigMap, so we couldn’t obtain the cluster ID. Therefore, I have two questions: First, what does this cluster ID mean? And how can I obtain the cluster ID for a manually created resource?

[kerenlahav]

Hi,
What is BTPO? are you migrating resources from svcat?

[nikoChen-I755695]

sry, it is BTPSO(BTP service operator). In fact, I'm migrating resources form another BTPSO and it is working. The problem I'm currently having is that I can't get the cluster ID when dealing with manually created resources. And it's not clear to me how cluster ID plays a role in the migration process. thx!

[kerenlahav]

When you deploy the btp operator you can define a cluster id, if not defined it will be generated.
While you can move resources between physical clusters, it is not possible to move resource between logical clusters (resource's cluster id cannot be changed). cluster id represents a logical cluster.

The link you shared is related to migrating resources from svcat environments to btp operator and is not relevant.
cluster id is one of the identifiers of a resource, in order to recover a resource it has to be with the same cluster id, in the same namespace and with the same k8s name as the original instance, cluster id of the resource is available in the cockpit (in instance details) or btpcli (in instance context).

what do you mean by "manually created resources"? each resource that was created with btp operator has a cluster id, you cannot recover resources that were not created with btp operator.

[nikoChen-I755695]

“cluster id of the resource is available in the cockpit (in instance details) or btpcli (in instance context).” https://github.com/SAP/sap-btp-service-operator/pull/466/files I tried to find the resource's cluster ID in Cockpit, but I couldn't seem to locate it. Could you provide a screenshot? Thank you very much!

The explanation of "manually created resources"
It can be understood as existing BTP resources. For example, resources that are created directly by the user through an interface. https://azure.github.io/azure-service-operator/guide/frequently-asked-questions/ To give a concrete example, I created a Redis through Azure portal, and then I can write an ASO CR to take over this Redis. Can the BTP Operator do that? If it is possible, please tell me exactly how to do it. If it is not possible, could you explain the reason? Thank you very much！

[kerenlahav]

If i understand you correctly you are trying to get the btp operator to manage resources that were not created by it, this is not supported.

[nikoChen-I755695]

Well, that’s unfortunate.

“cluster id of the resource is available in the cockpit (in instance details) or btpcli (in instance context).” https://github.com/SAP/sap-btp-service-operator/pull/466/files I tried to find the resource's cluster ID in Cockpit, but I couldn't seem to locate it. Could you provide a screenshot? Thank you very much!

I hope you can lend me a hand with this issue, thx!

[kerenlahav]

the cluster id is added by the btp operator, therefore you will not see it if the instance was not created with the it.

[nikoChen-I755695]

Extremely grateful！But, I discovered that the Cluster ID is stored in a ConfigMap called sap-btp-operator-config through the hack/setup_operator_env.sh script in the sap-btp-service-operator source code. However, when I checked this ConfigMap, I noticed that the Cluster ID there doesn’t match the one shown in Cockpit.



From the image, it seems that an instance has two different Cluster IDs. Could you help explain this?

[kerenlahav]

instance id is not cluster id, a cluster can have many instances (each instance has a different instance id), instance can be associated with one cluster only.
you can change the cluster id in the configmap and restart the operator after it, but if you already created instances with the existing cluster id it will lead to malfunction of the operator since you will have multiple cluster ids in the same cluster.

[nikoChen-I755695]

I'm sorry, I don't really understand. As shown in the images, the instance ID is consistently XXX-XXX-XXX-XXX-1ec84a361692, but the cluster ID in the ConfigMap is XXX-XXX-XXXX-XXX-45bcb5cb5944, while the cluster ID in Cockpit is XXX-XXX-XXXX-XXX-2b635841ccc4. These two cluster IDs do not match.

you can change the cluster id in the configmap and restart the operator after it, but if you already created instances with the existing cluster id it will lead to malfunction of the operator since you will have multiple cluster ids in the same cluster.

Yeah, I didn't do that. I did not change the configuration of the operator that created the instance

[kerenlahav]

instance's cluster id cannot be changed, the cluster id that you see in the cockpit is the correct one.
so it appears that the cluster's cluster id was changed by mistake or not.